Fuzzy Double Model Control for Air Supply on a PEM Fuel Cell System - Intelligent Computing for Sustainable Energy and Environment - page 398

e

r PID

N Z P

N PS PS PS

ec Z PB PS PB

P PS PS PS

Fig. 4. Fuzzy rules table of the fuzzy inference

N Z P PS P

1 1

0.5

0.5

0

0

0 0.2 0.4 0.6 0.8 1

-1 -0.5 0 0.5 1

Fig. 5. The membership functions of inputs and output in the fuzzy inference

T s =10 ms. The load current I load varies in a wide range from 20 A to 35 A and
load steps of different magnitudes were applied to the system. The compressor
voltage amplitude was restricted at the range of [0, 80].
From the Figure 6, it is clearly that the load current changes impacts on λ O 2 .
When the load current increases, the consumption of oxygen is corresponding
increase, and if the inlet mass of oxygen is constant, the oxygen excess ratio λ O 2
will decrease; whereas the λ O 2 will correspondingly increase in equal increments
as the consumption of oxygen if the load current decrease with the condition of
inlet oxygen is constant. From the work [15], we know that the output power
of fuel cell is the maximum power output when the λ O 2 is near 4. And if the
oxygen excess ratio λ O 2 is greater or less than 4, the output power of PEM fuel
cell is not the maximum. Even the starvation phenomenon would occur when the
oxygen excess ratio λ O 2 < 1. As can be seen from the simulation, it is necessary
to regulate the oxygen excess ratio at a suitable value.
It can be seen from the Figure 6 and 7 that the oxygen excess ratio is steadily
regulated at the setting value with the load current changed under the action

6

4

2

0

40

30

20

0 2.5 5 7.5 10 12.5 15 17.5

t (s)

Fig. 6. Oxygen excess ratio curves with load current changed