ONLINE OBSERVATION METHOD OF ANODE NITROGEN CONCENTRATION FOR PROTON EXCHANGE MEMBRANE FUEL CELL

Abstract

An online observation method of an anode nitrogen concentration for a proton exchange membrane fuel cell is disclosed. Firstly, a dynamic model of anode nitrogen concentration is established based on a gas transmembrane penetration model and an anode material conservation model of a fuel cell, and then an average voltage degradation value between a nitrogen partial pressure and an average monolithic cell voltage is obtained as online feedback information, an online observer of anode nitrogen concentration is established based on the dynamic model of anode nitrogen concentration and the online feedback information, and the anode nitrogen concentration of the fuel cell is obtained by the online observer. The new method solves the problem of online observation of anode nitrogen concentration during the operation of a proton exchange membrane fuel cell engine system under dynamic conditions.

Claims

1. An online observation method of an anode nitrogen concentration for a proton exchange membrane fuel cell, comprising: firstly, establishing a dynamic model of the anode nitrogen concentration based on a gas transmembrane penetration model and an anode material conservation model of the proton exchange membrane fuel cell, and then processing to obtain an average voltage degradation value between a nitrogen partial pressure and an average monolithic cell voltage as online feedback information, establishing an online observer of the anode nitrogen concentration based on the dynamic model of the anode nitrogen concentration and the online feedback information to obtain the anode nitrogen concentration of the proton exchange membrane fuel cell by the online observer.

2. The online observation method of the anode nitrogen concentration for the proton exchange membrane fuel cell according to claim 1, wherein the online observation method specifically comprises the following steps: 1) establishing the dynamic model of the anode nitrogen concentration: firstly, for a membrane electrode of a monolithic cell in the proton exchange membrane fuel cell, taking a penetration coefficient k.sub.N.sub.2 of the membrane electrode to nitrogen as a function ƒ(I.sub.st) of an output current I.sub.st of the proton exchange membrane fuel cell under conditions of a fixed temperature and a fixed gas humidity:
k.sub.N.sub.2=ƒ(I.sub.st) next, when the penetration coefficient of the membrane electrode to the nitrogen is known, calculating a nitrogen gas transmembrane penetration rate F.sub.N.sub.2 according to the penetration coefficient k.sub.N.sub.2, and a pressure difference between a cathode nitrogen partial pressure P.sub.N.sub.2.sub.,ca and an anode nitrogen partial pressure P.sub.N.sub.2.sub.,an of the proton exchange membrane fuel cell:
F.sub.N.sub.2=k.sub.N.sub.2(P.sub.N.sub.2.sub.,ca−P.sub.N.sub.2.sub.,an) then, establishing the following dynamic model of the anode nitrogen concentration according to an operating temperature T.sub.an of an anode of the proton exchange membrane fuel cell, a gas constant parameter R, and a volume V.sub.an of an anode loop of the proton exchange membrane fuel cell: $P_{N_2,\mathrm{an}}=\frac{{\mathrm{RT}}_{\mathrm{an}}}{V_{\mathrm{an}}}{F}_{N_2}$ 2) obtaining an ideal value of the average monolithic cell voltage between the nitrogen partial pressure and an average voltage according to the dynamic model of the anode nitrogen concentration: $V_{\mathrm{a\nu g}\mathrm{cell}}=E_0+b_0\left(\ln \frac{P_{\mathrm{an}}-P_{N_2,\mathrm{an}}}{P_0}+\frac{1}{2}\ln \frac{P_{c,O_2}}{P_0}\right)-v_{\mathrm{ohm}}-v_{\mathrm{act}}-v_{\mathrm{con}}$ wherein, V.sub.avgcell denotes the ideal value of the average monolithic cell voltage, E.sub.0 denotes a reversible voltage of the monolithic cell in an ideal state, b.sub.0 denotes a gas pressure parameter, P.sub.an denotes an anode pressure, P.sub.N.sub.2an denotes the anode nitrogen partial pressure, P.sub.c,O.sub.2 denotes a cathode oxygen partial pressure, P.sub.0 denotes atmospheric pressure, V.sub.ohm denotes an ohmic loss voltage of the monolithic cell, V.sub.act denotes a polarization loss voltage of the monolithic cell, and v.sub.con denotes a concentration loss voltage of the monolithic cell; obtaining the average voltage degradation value ΔV.sub.avgcell between the anode nitrogen partial pressure and the average monolithic cell voltage according to a difference between a reference monolithic cell voltage with the anode nitrogen concentration of 0 and the average monolithic cell voltage in a presence of oxygen and the nitrogen concentration P.sub.N.sub.,an: $\Delta V_{\mathrm{a\nu g}\mathrm{cell}}=b_0\ln \frac{P_{\mathrm{an}}-P_{N_2,\mathrm{an}}}{P_0}$ 3) establishing the following online observer of the anode nitrogen concentration to obtain the anode nitrogen concentration of the proton exchange membrane fuel cell in real time by an observation of the online observer: ${\hat{\alpha }}_{N_2,\mathrm{an}}=c_1-c_2{\alpha }_{N_2,\mathrm{an}}+H\left(y\left(t\right)-\Delta V_{\mathrm{a\nu g}\mathrm{cell}}\right)$ $c_2=\frac{R{T}_{\mathrm{an}}{k}_{N_2}}{V_{\mathrm{an}}},c_1=\frac{R{T}_{\mathrm{an}}{k}_{N_2}{P}_{N_2,\mathrm{ca}}}{V_{\mathrm{an}}{P}_{\mathrm{an}}},{\hat{\alpha }}_{N_2,\mathrm{an}}=\frac{P_{N_2,\mathrm{an}}}{P_0}$ wherein, c.sub.2 and c.sub.1 denote a first constant parameter and a second constant parameter of the online observer, respectively, â.sub.N.sub.2.sub.,an denotes an observed value of the anode nitrogen concentration; H denotes a gain parameter of the online observer, ΔV.sub.avgcell denotes the average voltage degradation value, and y(t) denotes a measured average monolithic cell voltage degradation value, and the y(t) is calculated by the following formula:
y(t)=V(t)−V* wherein, V(t) denotes a current average monolithic cell voltage measured value, and V* denotes an ideal value of an average monolithic voltage under a current operation condition.

3. The online observation method of the anode nitrogen concentration for the proton exchange membrane fuel cell according to claim 2, wherein a current of the proton exchange membrane fuel cell undergoes a step change or the anode of the proton exchange membrane fuel cell undergoes a purge for a transition time of t, and after a voltage measured value is stable, the current average monolithic cell voltage measured value V(t) after stabilization is selected as the ideal value of the average monolithic voltage under the current operation condition.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0025] FIG. 1 shows a method for selecting an ideal value of an average monolithic voltage under a current operation condition for an observer in the present invention.

[0026] FIG. 2 shows results of experimental identification of nitrogen penetration coefficients according to an embodiment of the present invention.

[0027] FIG. 3 shows a change curve of fuel cell outputs and gas concentrations within a purge cycle according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] The present invention is further described below with reference to the accompanying drawings and embodiments.

[0029] According to the method for implementing a nitrogen concentration observer described in summary of the present invention, the observer is realized in a computer program of an experimental bench.

[0030] 1) The dynamic model of anode nitrogen concentration is established.

[0031] Firstly, for a membrane electrode of a monolithic cell in the fuel cell, a penetration coefficient k.sub.N.sub.2 of the membrane electrode to nitrogen is taken as a function ƒ(I.sub.st) of an output current I.sub.st of the fuel cell under conditions of a fixed temperature and a fixed gas humidity.

[0032] Next, when the penetration coefficient of the membrane electrode to nitrogen is known, a nitrogen gas transmembrane penetration rate F.sub.N.sub.2 is calculated according to the penetration coefficient k.sub.N.sub.2, and a pressure difference between a cathode nitrogen partial pressure P.sub.N.sub.2.sub.,ca and an anode nitrogen partial pressure P.sub.N.sub.2.sub.,an of the fuel cell.

[0033] Then, the following dynamic model of the anode nitrogen concentration is established according to an operating temperature T.sub.an of an anode of the fuel cell, a gas constant parameter R, and a volume V.sub.an of an anode loop of the fuel cell.

[0034] 2) An ideal value of a monolithic cell voltage between the nitrogen partial pressure and the average voltage is obtained according to the dynamic model of anode nitrogen concentration; then, the average voltage degradation value ΔV.sub.avgcell between the anode nitrogen partial pressure and the average monolithic cell voltage is obtained according to a difference between a reference monolithic cell voltage with the anode nitrogen concentration of 0 and the monolithic cell voltage in the presence of oxygen and the nitrogen concentration P.sub.N.sub.2.sub.,an.

[0035] 3) The following online observer of the anode nitrogen concentration is established, and the anode nitrogen concentration of the fuel cell is obtained in real time by an observation of the online observer.

[0036] In specific implementation, the current of the fuel cell undergoes a step change or the anode of the fuel cell undergoes a purge for a transition time of t, and after the voltage measured value is stable, a stable average monolithic voltage measured value V(t) is selected as the ideal value of the average monolithic voltage under the current operation condition.

[0037] In order to obtain a penetration coefficient of nitrogen in a membrane electrode, an experiment is conducted to identify the penetration coefficient of nitrogen in the membrane electrode. When other conditions remain unchanged, the penetration coefficient of nitrogen varies linearly with the current density in a working range, and a changing curve is estimated by obtaining experimental values of two points. Under the current density of 0.4 A/cm.sup.2, the penetration coefficient of nitrogen is identified as 1.490×10.sup.−9 mol/(Pa.Math.s). Under the current density of 0.6 A/cm.sup.2, the penetration coefficient of nitrogen is identified as 1.994×10.sup.−9 mol/(Pa.Math.s).

[0038] Then, the observer provided in the present invention is realized in a controller. During the experiment, a constant metering ratio of 2.5 is set for the cathode, the oxygen content in the test environment is 21%, the anode pressure and the cathode pressure are set according to reference values of a fuel cell manufacturer, a fuel cell coolant inlet temperature is controlled at 60° C.±0.5° C., and a fuel cell coolant outlet temperature is controlled at 65° C.±0.5° C.

Implementation results of examples are as shown in FIG. 3. In FIG. 3, in one purge cycle, step changes of load are carried out several times, and a reference voltage is re-selected based on the algorithm in FIG. 1 in each change process. The whole purge cycle is not affected by load changes, and finally the nitrogen concentrations under the whole working condition are observed continuously.