0
Research Papers

# Enhancing Polymeric Electrolyte Membrane Fuel Cell Control by Means of the Perturb and Observe Technique

[+] Author and Article Information
A. Giustiniani

Department of Electrical and Information Engineering, Università di Salerno, Fisciano, Salerno 84084, Italyagiustiniani@unisa.it

G. Petrone

Department of Electrical and Information Engineering, Università di Salerno, Fisciano, Salerno 84084, Italygpetrone@unisa.it

G. Spagnuolo

Department of Electrical and Information Engineering, Università di Salerno, Fisciano, Salerno 84084, Italygspagnuolo@unisa.it

I. Arsie

Department of Mechanical Engineering, Università di Salerno Fisciano, Salerno 84084, Italyiarsie@unisa.it

A. Di Domenico

Department of Mechanical Engineering, Università di Salerno Fisciano, Salerno 84084, Italyadidomenico@unisa.it

C. Pianese

Department of Mechanical Engineering, Università di Salerno Fisciano, Salerno 84084, Italypianese@unisa.it

M. Sorrentino

Department of Mechanical Engineering, Università di Salerno Fisciano, Salerno 84084, Italymsorrentino@unisa.it

M. Vitelli

Department of Information Engineering, Seconda Università di Napoli Aversa (CE) 81031, Italyvitelli@unina.it

J. Fuel Cell Sci. Technol 7(1), 011021 (Nov 11, 2009) (11 pages) doi:10.1115/1.3120275 History: Received February 17, 2008; Revised July 21, 2008; Published November 11, 2009; Online November 11, 2009

## Abstract

In this paper, the use of an adaptive technique aimed at controlling a polymeric electrolyte membrane fuel cell is introduced. It is demonstrated that a hill climbing-based method, acting on the compressor speed and/or the cathode back-pressure valve, allows to better take into account the effect of exogenous variables on stack performance. Particularly, the proposed technique has proven to perform better than classical feedforward/feedback approaches when well known aging mechanisms deteriorate cell efficiency. Numerical results based on experimentally derived models confirm the potential of the proposed control method and its intrinsic reliability.

<>

## Figures

Figure 1

Layout of PEM FCS. In the figure the supply and exhaust manifolds and the electrode volumes are omitted for the sake of simplicity.

Figure 2

Comparison between experiments and model estimations for a FC stack at different operating pressures. Cell temperature set to 94°C.

Figure 3

Figure 4

Dynamic system with the perturbed parameter p and the observed performance function f(p)

Figure 5

The climbing of the performance function f by means of repeated variations in the perturbed parameter p

Figure 6

FCS efficiency versus stack current

Figure 7

Circuitry for FCS current control

Figure 8

Stepwise increasing FCS efficiency obtained by means of P&O acting on the output current: (a) complete transient and (b) magnification of the same waveform at steady state

Figure 9

Stepwise waveforms of the time domain transient leading to maximum efficiency conditions: (a) FCS current, (b) FCS voltage, and (c) excess air variation

Figure 10

Perturbation of the electro-active area used for simulations (nominal value of 300 cm2)

Figure 11

Control system layout: adaptive control (P&O)-based correction of the compressor speed

Figure 12

Power electronic circuit

Figure 13

FCS output power. Bold line represents the target output power, dashed line represents the output power obtained by using lookup tables only, and solid line represents the output power obtained by the P&O-based trimming of the compressor speed.

Figure 14

Stepwise waveform of the perturbed compressor speed ωcmd. The thin line represents the constant value of ωcmd imposed by using the lookup tables only.

Figure 15

Stepwise waveform of the excess air variation. The thin line is the constant value expected (but not really obtained) by using the lookup tables only.

Figure 16

Waveforms of the FCS pressures

Figure 17

FCS net output power. Bold line represents the target output power, dashed line represents the output power obtained by reducing the back-pressure valve opening, and solid line represents the output power obtained by the P&O-based trimming of the compressor speed.

Figure 18

Compressor speed imposed by the P&O technique. Thin line represents the reduced back-pressure valve opening and thick line represents the (P&O)-based trimming of the compressor speed.

Figure 19

Two variable control system layout: adaptive control (P&O)-based correction of the compressor speed and of the back-pressure valve aperture

Figure 20

Perturbation of the electro-active area used for simulations (nominal value of 300 cm2)

Figure 21

Net power at the FCS output. Continuous line represents the P&O method, dashed thin line represents the use of lookup tables only, and dashed thick line represents the target power level.

Figure 22

Power at the stack output. Continuous line represents the P&O method and dashed line represents the use of lookup tables only.

Figure 23

An example of steady state conditions under two-variable perturbation: (a) time domain and (b) control variable domain

Figure 24

(a) Cathode pressure and (b) excess air variation during the transient. Pressure variations occur every 10 s.

Figure 25

Net power at the FCS output. Continuous thin line represents the P&O method, dashed thin line represents the use of lookup tables only, and dashed thick line represents the target power level.

Figure 26

Power at the stack output. Continuous line represents the P&O method and dashed line represents the use of lookup tables only.

Figure 27

An example of steady state conditions under two-variable adaptive perturbation: (a) time domain and (b) control variable domain

Figure 28

Cathode pressure and excess air variation during the transient

## Discussions

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections