06 Electrical power supply and utilization (scientific, technical)

04/02323
Reactor

ProbabUistic safety analysis of a Greek Research

Aneziris, O. N. et al. Annals of Nuclear Energy, 2004, 31, (5), 481-516.

This paper presents the work and the results of a Level 1 Probabilistic
Safety Assessment performed for the Greek Research Reactor of the
National Centre for Scientific Research 'Demokritos'. Event trees have
been used to study the response of the installation to various initiating
events whereas fault trees have been used in the modelling of safety
system failures. Using generic data, the probability of an accident
leading to severe core damage (_>10% of core melted) from internal
initiating events has been estimated to be 310 .6 per year of reactor
operatiom The largest contributions to the probability of small damage
are made by accidents initiated by loss of cooling and excess reactivity.
Large releases of radioactivity are expected with a negligibly low
frequency of about 1 xl0~ l~/year.

04102324 Trends in computing systems for large fusion

experiments
How, J. A. et al. Fusion Engineering and Design, 2004, 70, (2), 115-122.
The major European fusion facilities, JET, Tore Supra and Asdex
Upgrade, have a 30-year history from conception to the present day.
This period has seen a rapid evolution in computer techniques.
Experience has shown that IT architectures and standards adopted
early in the construction phase of a project are carried into the main
programme and affect the project for its entire lifetime. This paper
selects a few of the more important computer-related lessons learned
in Europe, particularly with respect to the EFDA-JET facility, and
makes recommendations towards the next generation of machines.

06

ELECTRICAL POWER
SUPPLY A N D
UTILIZATION
Scientific, technical

04/02325 A flexible approach for overcurrent relay

characteristics simulation
Karegar, H. K. et at. Electric Power Systems Research, 2003, 66, (3L
233-239.
Accurate models of overcurrent (OC) with inverse time relay
characteristics play an important role for the coordination of power
system protection schemes. This paper proposes a new method for
modelling OC relays curves. The model is based on fuzzy logic and
artificial neural network. The feedforward multilayer perceptron
neural network is used to calculate the operating times of OC relays
for various time dial settings (TDS) or time multiplier settings (TMS).
The new model is more accurate than traditional models. The model
has been validated by comparing the results obtained from the new
method with linear and non-linear analytical models.

04102326 A global thermo-electrochemical model for SOFC

systems design and engineering
Petruzzi, L. et al. Journal of Power Sources, 2003, 118, (1-2), 96-107.
At B M W AG in Munich high-temperature solid oxide fuel cells
(SOFCs) are being developed as an auxiliary power unit (APU) for
high-class car conveniences. Their design requires simulation of their
thermo-electrochemical behaviour in all the conditions that may occur
during operation (i.e. heat-up to about 600C, start-up to operating
temperature, energy-delivering and cool-down). A global thermoelectrochemical model was developed for the whole system and a threedimensional geometry code was performed using MATLAB programming language. The problems in developing SOFCs are now so many
and so different that a very flexible code is necessary. Thus, the code
was not only designed in order to simulate each of the operating
conditions, but also to test different stack configurations, materials, etc.
In every event, the code produces a time-dependent profile of
temperatures, currents, electrical and thermal power density, gases
concentrations for the whole system. The heat-up and start-up
simulations allow: (1) to evaluate the time the cell stack needs to
reach operating temperature from an initial temperature distribution,
(2) to check the steepest temperature gradients occurring in the
ceramic layers (which result in material stresses) and (3) to obtain
important information about the pre-operating strategy. Simulation of
energy-dellvering gives a detailed profile of the temperatures, currents,
power density,, and allows to define the guidelines in system-

controlling. Simulation of cooling-down gives important advises about

insulation designing. The aim of this work is to build up a tool to clearly
individuate the best designing criteria and operating strategy during
the development and the engineering of a SOFC system.

04102327 A two-phase flow model for hydrogen evolution in

an electrochemical cell
Mat, M. D. M. D. el al. International Journal of Hydrogen Energy, 2004,
29, (~0), 1015-1023.
Hydrogen evolution, flow field and current density distribution in an
electrochemical cell are investigated with a two-phase flow model. The
mathematical model involves solutions of transport equations for the
variables of each phase with allowance for inter-phase transfer of mass
and momentum. The buoyancy force generated due to density
difference between two phases modifies flow profile and increases
fluid velocity at the vicinity of the electrode. The current density
decreases over the electrode mainly because of the decrease in effective
conductivity of electrolyte. It is found that the hydrogen generation
significantly increases at higher electrolyte flow by reducing the
residence time of bubbles over the electrode. The predicted results
satisfactorily agree with data available in the literature.

04/02328

Bipolar plates for PEM fuel cells

Middelman, E. et al. Journal of Power Sources, 2003, 118, (1 2), 44-46.

The bipolar plates are in weight and volume the major part of the PEM
fuel cell stack, and are also a significant contributor to the stack costs.
The bipolar plate is therefore a key component if power density has to
increase and costs must come down. Three cell plate technologies are
expected to reach targeted cost price levels, all having specific
advantages and drawbacks. NedStack has developed a conductive
composite materials and a production process for fuel cell plates
(bipolar and mono-polar). The material has a high electric and thermaI
conductivity, and can be processed into bipolar plates by a proprietary
molding process. Process cycle time has been reduced to less than 10 s,
making the material and process suitable for economical mass
production. Other development work to increase material efficiency
resulted in thin bipolar plates with integrated cooling channels, and
integrated seals, and in two-component bipolar plates. Total thickness
of the bipolar plates is now less than 3 ram, and will be reduced to 2
m m in the near future. With these thin integrated plates it is possible to
increase power density up to 2 kW/1 and 2 kW/kg, while at the same
time reducing cost by integrating other functions and less material use.

04/02329 Capacitance of the polypyrrole/polyimide

composite by electrochemical impedance spectroscopy
Iroh, J. O. and Levine, K. Journal o/'Poa,er Sources, 2003, 117, (1-2),
267-27Z
Electrically conducting polymers and their composites attract a lot of
attention because of their high charge storage ability. As was shown
previously, the composite of polypyrrole (PPy) and polyimide (PI)
possesses better charge storage properties then pure PPy. Probably the
PI matrix protects PPy from oxidative destruction and serves as a high
molecular weight dopant. In this paper, the PPy/PI composite was
studied by as electrochemical impedance spectroscopy at different
applied dc polarization potentials. The properties of the coatings were
found to differ significantly between pure PI and the PPy/P] composite,
depending on the applied potential and on the amount of the
conducting polymer. An additional time constant related to the PPy/
PI interface appears when PPy is added to PI. The pseudocapacitance
of the PPy/PI composite increased significantly with increased cathodic
potential. This was explained by additional doping of PPy by the PI
matrix.

04/02330 Capacity fade study of lithium-ion batteries

cycled at high discharge rates
Ning, G. et al. Journal of Power Sources', 2003, 1 t7, (1-2), 160-169.
Capacity fade of Sony US 18650 Li-ion batteries cycled using different
discharge rates was studied at ambient temperature. The capacity
losses were estimated after 300 cycles at 2C and 3C discharge rates and
were found to be 13.2 and 16.9% of the initial capacity, respectively. At
1C discharge rate the capacity lost was only 9.5%. The cell cycled at
high discharge rate (3C) showed the largest internal resistance increase
of 27.7% relative to the resistance of the fresh cells. The rate capability
losses were proportional with the increase of discharge rates. Half-cell
study and material and charge balances were used to quantify the
capacity fade due to the losses of primary active material (El +), the
secondary active material (LiCoOz/C)) and rate capability losses. It
was found that carbon with 10.6% capacity loss after 300 cycles
dominates the capacity fade of the whole cell at high discharge rates
(3C). A mechanism is proposed which explains the capacity fade at high
discharge rates.

Fuel and Energy Abstracts

September 2004 329