0
RESEARCH PAPERS

Dynamic Models for Hydrogen Peroxide Control in Solar Photo-Fenton Systems

[+] Author and Article Information
J. D. Alvarez

CIEMAT—Plataforma Solar de Almería, P.O. Box 22, 04200 Tabernas, University of Almería, Dpto. de Lenguajes y Computación, Área de Ingeniería de Sistemas y Automática, Crta. Sacramento s/n, 04120 La Cañada, Almería, Spain

W. Gernjak

CIEMAT—Plataforma Solar de Almería, P.O. Box 22, 04200 Tabernas, Almería, Spain, University of Applied Life Sciences, Department of Water, Air and Environment, Muthgasse 18, 1190 Vienna, Austria

S. Malato

CIEMAT—Plataforma Solar de Almería, P.O. Box 22, 04200 Tabernas, Almería, Spain

M. Berenguel

Dpto. de Lenguajes y Computación, Área de Ingeniería de Sistemas y Automática, Crta. Sacramento s/n, 04120 La Cañada, Almería, Spain

M. Fuerhacker

University of Applied Life Sciences, Department of Water, Air and Environment, Muthgasse 18, 1190 Vienna, Austria

L. J. Yebra

CIEMAT—Plataforma Solar de Almería, P.O. Box 22, 04200 Tabernas, Almería, Spain, Dpto. de Lenguajes y Computación, Área de Ingeniería de Sistemas y Automática, Crta. Sacramento s/n, 04120 La Cañada, Almería, Spainluis.yebra@psa.es

J. Sol. Energy Eng 129(1), 37-44 (Dec 10, 2005) (8 pages) doi:10.1115/1.2391014 History: Received July 22, 2005; Revised December 10, 2005

Abstract

Simulations and real decomposition experiments with hydrogen peroxide induced by dissolved iron and solar illumination were performed in a solar pilot plant with compound parabolic collectors designed for photo-Fenton wastewater treatment. The structure of a gray-box linear model aimed at reproducing system dynamics with parameters dependent on operating conditions when operating around prescribed set-points is proposed. This simple model relates the changes in hydrogen peroxide concentration due to changes in hydrogen peroxide injection, dissolved iron, and solar illumination. Based on this model, control of the hydrogen peroxide concentration at set points between $200–900mg∕L±10mg∕L$ under dynamic conditions (simultaneous decomposition and addition of hydrogen peroxide by a dosage pump), was achieved. Different basic approaches for controller actuation upon the frequency of the dosage pump were tested, ranging from a simple PI controller to a PI controller plus antiwindup action and feedforward control. From these basic approaches, conclusions can be drawn about the process’s behavior under closed-loop control.

<>

Figures

Figure 1

Flow scheme of pilot plant

Figure 2

System response to an input change. Wide solid line: percentage of H2O2 pump frequency range, thin solid line: simulated output model, dashed line: real H2O2 concentration, (a) pulse of 5min duration with H2O2 pump operating at maximum frequency, (b) pulse of 5min duration with H2O2 pump operating at 20% of its maximum frequency

Figure 3

System consumption in the presence of Fe, H2O2. Solid line: simulated output model, dashed line: real H2O2 concentration, (a) cFe=1.43mg∕L, collectors were exposed to sunlight from 9060s<t<10260s. The rest of the time the collector was not illuminated; (b) cFe=0.63mg∕L, collectors were exposed to sunlight from 5580s<t<6780s. The rest of the time the collector was not illuminated

Figure 4

PI controller+antiwindup

Figure 5

Control of CH2O2: (a) system output; (b) controller output to the dosage pump, (c) real percentage of H2O2 pump frequency. Solid lines: PI controller with anti-windup; dashed lines: PI controller without antiwindup

Figure 6

PI controller+antiwindup+feedforward control

Figure 7

System output controlled by a PI with antiwindup (solid line) including a feedforward module less aggressive (dashed line) and including a feedworward module more aggressive (dotted line)

Figure 8

Real test (a) system output controlled with a PI with antiwindup; (b) controller output to dosage pump

Figure 9

Real test, (a) real system output controlled by a PI with antiwindup and feedforward module (dotted line) and simulated output model (dashed line); (b) real controller output to the dosage pump (dotted line) and simulated controller output (dashed line). The collector was exposed to illumination all the time. The system setpoint was fixed to 500mg∕L, as indicated by the solid line in (a)

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