Prediction of Direct Methanol Fuel Cell Stack Performance Using Artificial Neural Network

2017 ◽  
Vol 14 (3) ◽  
Author(s):  
M. A. Rafe Biswas ◽  
Melvin D. Robinson
Keyword(s):  
Fuel Cell ◽  
Direct Methanol Fuel Cell ◽  
Experimental Studies ◽  
Cell Voltage ◽  
Operating Conditions ◽  
Coefficient Of Determination ◽  
Membrane Electrode ◽  
Direct Methanol ◽  
Artificial Neural ◽  
Methanol Fuel Cell

A direct methanol fuel cell (DMFC) converts liquid fuel into electricity to power devices, while operating at relatively low temperatures and producing virtually no greenhouse gases. Since DMFC performance characteristics are inherently complex, it can be postulated that artificial neural networks (NN) represent a marked improvement in prediction capabilities. In this work, an artificial NN is employed to predict the performance of a DMFC under various operating conditions. Input variables for the analysis consist of methanol concentration, temperature, current density, number of cells, and anode flow rate. The addition of the two latter variables allows for a more distinctive model when compared to prior NN models. The key performance indicator of our NN model is cell voltage, which is an average voltage across the stack and ranges from 0 to 0.8 V. Experimental studies were conducted using DMFC stacks with membrane electrode assemblies consisting of an additional unique liquid barrier layer to minimize water loss to atmosphere. To determine the best fit to the experimental data, the model is trained using two second-order training algorithms: OWO-Newton and Levenberg–Marquardt (LM). The topology of OWO-Newton algorithm is slightly different from that of LM algorithm by employing bypass weights. The application of NN shows rapid construction of a predictive model of cell voltage for varying operating conditions with an accuracy on the order of 10−4, which can be comparable to literature. The coefficient of determination of the optimal model results using either algorithm were greater than 0.998.

Performance Estimation of Direct Methanol Fuel Cell Using Artificial Neural Network

2015 ◽  
Author(s):  
M. A. Rafe Biswas ◽  
Melvin D. Robinson
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Fuel Cell ◽  
Direct Methanol Fuel Cell ◽  
Cell Voltage ◽  
Operating Conditions ◽  
Direct Methanol ◽  
Artificial Neural ◽  
Methanol Fuel Cell

A direct methanol fuel cell can convert chemical energy in the form of a liquid fuel into electrical energy to power devices, while simultaneously operating at low temperatures and producing virtually no greenhouse gases. Since the direct methanol fuel cell performance characteristics are inherently nonlinear and complex, it can be postulated that artificial neural networks represent a marked improvement in performance prediction capabilities. Artificial neural networks have long been used as a tool in predictive modeling. In this work, an artificial neural network is employed to predict the performance of a direct methanol fuel cell under various operating conditions. This work on the experimental analysis of a uniquely designed fuel cell and the computational modeling of a unique algorithm has not been found in prior literature outside of the authors and their affiliations. The fuel cell input variables for the performance analysis consist not only of the methanol concentration, fuel cell temperature, and current density, but also the number of cells and anode flow rate. The addition of the two typically unconventional variables allows for a more distinctive model when compared to prior neural network models. The key performance indicator of our neural network model is the cell voltage, which is an average voltage across the stack and ranges from 0 to 0:8V. Experimental studies were carried out using DMFC stacks custom-fabricated, with a membrane electrode assembly consisting of an additional unique liquid barrier layer to minimize water loss through the cathode side to the atmosphere. To determine the best fit of the model to the experimental cell voltage data, the model is trained using two different second order training algorithms: OWO-Newton and Levenberg-Marquardt (LM). The OWO-Newton algorithm has a topology that is slightly different from the topology of the LM algorithm by the employment of bypass weights. It can be concluded that the application of artificial neural networks can rapidly construct a predictive model of the cell voltage for a wide range of operating conditions with an accuracy of 10−3 to 10−4. The results were comparable with existing literature. The added dimensionality of the number of cells provided insight into scalability where the coefficient of the determination of the results for the two multi-cell stacks using LM algorithm were up to 0:9998. The model was also evaluated with empirical data of a single-cell stack.

scholarly journals Performance of passive Direct Methanol Fuel Cell: modelling and experimental studies

2017 ◽  
Vol 1 (1) ◽  
pp. 89-103 ◽  
Author(s):  
Beatriz A. Berns ◽  
Mariana F. Torres ◽  
Vânia B. Oliveira ◽  
Alexandra M. F. R. Pinto
Keyword(s):  
Fuel Cell ◽  
Direct Methanol Fuel Cell ◽  
Experimental Studies ◽  
Gas Diffusion ◽  
Membrane Electrode ◽  
Membrane Area ◽  
Diffusion Layers ◽  
Electrode Assemblies ◽  
Direct Methanol ◽  
Methanol Fuel Cell

Low methanol and water crossover with high methanol concentrations are essential requirements for a passive Direct Methanol Fuel Cell (DMFC) to be used in portable applications. Therefore, it is extremely important to clearly understand and study the effect of the different operating and configuration parameters on the cell’s performance and both methanol and water crossover. In the present work, a detailed experimental study on the performance of an in-house developed passive DMFC with 25 cm2 of active membrane area is described. Tailored membrane electrode assemblies (MEAs) with different structures and combinations of gas diffusion layers (GDL) and membranes, were tested in order to select optimal working conditions at high methanol concentration levels without sacrificing performance. The experimental polarization curves were successfully compared with the predictions of a steady state, one-dimensional model accounting for coupled heat and mass transfer, along with the electrochemical reactions occurring in the passive DMFC developed by the same authors.

Modeling of Direct Methanol Fuel Cell Using the Artificial Neural Network

2013 ◽  
Vol 10 (4) ◽  
Author(s):  
Mehdi Tafazoli ◽  
Hamid Baseri ◽  
Ebrahim Alizadeh ◽  
Mohsen Shakeri
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Fuel Cell ◽  
Direct Methanol Fuel Cell ◽  
Cell Voltage ◽  
Neural Network Models ◽  
Input Parameters ◽  
Direct Methanol ◽  
Artificial Neural ◽  
Methanol Fuel Cell

The performance of a direct methanol fuel cell (DMFC) has complex nonlinear characteristics. In this paper, the performance of a DMFC has been modeled using a neural network approach. The input parameters of the DMFC model include cell geometrical and operational parameters such as the cell temperature, oxygen flow rate, channel depth of the bipolar plate, methanol concentration, cathode back pressure, and current density and the output parameter is the cell voltage. In order to predict the performance of a DMFC single cell, two types of artificial neural network (ANN) have been developed to correlate the input parameters of the DMFC to the cell voltage. The performance of the networks was investigated by varying the number of neurons, number of layers, and transfer function of the ANNs and the best one is selected based on the mean square error. The results indicated that the neural network models can predict the cell voltage with an acceptable accuracy.

3D Numerical Study of a Flowing Electrolyte – Direct Methanol Fuel Cell With an Implemented Bi-Layered Membrane Electrode Diaphragm Assembly

2011 ◽  
Author(s):  
P. A. Cornellier ◽  
E. Matida ◽  
C. A. Cruickshank
Keyword(s):  
Fuel Cell ◽  
Flow Rate ◽  
Direct Methanol Fuel Cell ◽  
Numerical Study ◽  
Experimental Studies ◽  
Membrane Electrode ◽  
Pressure Losses ◽  
Direct Methanol ◽  
Flow Through ◽  
Methanol Fuel Cell

In the present work, fluid dynamic simulation and experimental studies are compared to assess the validity of using computational fluid dynamics (CFD) to accurately predict the pressure losses experienced across each of the three fluid channels in a flowing electrolyte direct methanol fuel cell: methanol flow through anodic-serpentine channels; air flow through the cathodic-serpentine channels; dilute sulfuric acid flow through the flowing electrolyte (FE) channel located between two membrane-electrode assemblies (MEAs). The methanol flow rate is varied from 5 to 25 mL/min and the airflow is varied from 0.5 to 5 L/min. The flowing electrolyte flow rate is also varied from 5 to 25 mL/min in order to analyze pressure levels within the FE channel, which, according to this analysis, must be larger than the adjacent serpentine channels. This pressure difference is particularly important to maintain the distance (and flow structure) between the MEAs without affecting performance of the fuel cell. Adequately controlling the pressure of each of three fluids disables the MEAs ability to deform without the use of an electrolyte spacer, effectively creating an inter-dependent bi-layered membrane electrode diaphragm assembly (Bi-MEDA). Through CFD simulation, it was observed that pressure equalization through the Bi-MEDA approach supports the elimination of a flowing electrolyte channel spacer from current FE-DMFC designs. The reduction of the spacer is expected to decrease ohmic losses currently experienced in all FE-DMFC designs. Despite several approximations, simulations predicting pressure losses throughout the two serpentine fuel channels are compared against obtained experimental data, showing relatively good agreement for a single cell arrangement.

Experimental Studies of a Formic Acid Electrolyte: Direct Methanol Fuel Cell

2013 ◽  
Author(s):  
David Ouellette ◽  
Cynthia Ann Cruickshank ◽  
Edgar Matida
Keyword(s):  
Fuel Cell ◽  
Formic Acid ◽  
Power Output ◽  
Direct Methanol Fuel Cell ◽  
Experimental Studies ◽  
Operating Conditions ◽  
Acid Electrolyte ◽  
Direct Methanol ◽  
A Cell ◽  
Methanol Fuel Cell

A new methanol fuel cell that utilizes a liquid formic acid electrolyte, named the formic acid electrolyte-direct methanol fuel cell (FAE-DMFC) is experimentally tested. Three fuel cell configurations were examined; a flowing electrolyte and two circulating electrolyte configurations. From these three configurations, the flowing electrolyte and the circulating electrolyte, with the electrolyte outlet routed to the anode inlet, provided the most stable power output; where minimal decay in performance and less than 3 and 5.6 % variation in power output were observed in the respective configurations. The flowing electrolyte configuration also yielded the greatest power output by as much as 34 %. Furthermore, for the flowing electrolyte configuration, several key operating conditions were experimentally tested to determine the optimal operating points. It was found that an inlet concentration of 2.2 M methanol and 6.5 M formic acid, as well as a cell temperature of 52.8 °C provided the best performance.

Experimental Investigation on the Performance of a Formic Acid Electrolyte-Direct Methanol Fuel Cell

2013 ◽  
Vol 11 (2) ◽  
Author(s):  
David Ouellette ◽  
Cynthia Ann Cruickshank ◽  
Edgar Matida
Keyword(s):  
Fuel Cell ◽  
Formic Acid ◽  
Power Output ◽  
Direct Methanol Fuel Cell ◽  
Operating Conditions ◽  
Portable Devices ◽  
Acid Electrolyte ◽  
Optimal Operating ◽  
Direct Methanol ◽  
Methanol Fuel Cell

The performance of a new methanol fuel cell that utilizes a liquid formic acid electrolyte, named the formic acid electrolyte-direct methanol fuel cell (FAE-DMFC) is experimentally investigated. This fuel cell type has the capability of recycling/washing away methanol, without the need of methanol-electrolyte separation. Three fuel cell configurations were examined: a flowing electrolyte and two circulating electrolyte configurations. From these three configurations, the flowing electrolyte and the circulating electrolyte, with the electrolyte outlet routed to the anode inlet, provided the most stable power output, where minimal decay in performance and less than 3% and 5.6% variation in power output were observed in the respective configurations. The flowing electrolyte configuration also yielded the greatest power output by as much as 34%. Furthermore, for the flowing electrolyte configuration, several key operating conditions were experimentally tested to determine the optimal operating points. It was found that an inlet concentration of 2.2 M methanol and 6.5 M formic acid, as along with a cell temperature of 52.8 °C provided the best performance. Since this fuel cell has a low optimal operating temperature, this fuel cell has potential applications for handheld portable devices.

Critique and Improvement of a One-Dimensional Semianalytical Model of a Direct Methanol Fuel Cell

2012 ◽  
Vol 9 (5) ◽  
Author(s):  
C. C. Kuo ◽  
W. E. Lear ◽  
J. H. Fletcher ◽  
O. D. Crisalle
Keyword(s):  
Fuel Cell ◽  
Polarization Curve ◽  
Current Density ◽  
Direct Methanol Fuel Cell ◽  
Membrane Electrode Assembly ◽  
Membrane Electrode ◽  
One Dimensional ◽  
Implicit Equation ◽  
Direct Methanol ◽  
Methanol Fuel Cell

A constructive critique and a suite of proposed improvements for a recent one-dimensional semianalytical model of a direct methanol fuel cell are presented for the purpose of improving the predictive ability of the modeling approach. The model produces a polarization curve for a fuel cell system comprised of a single membrane-electrode assembly, based on a semianalytical one-dimensional solution of the steady-state methanol concentration profile across relevant layers of the membrane electrode assembly. The first improvement proposed is a more precise numerical solution method for an implicit equation that describes the overall current density, leading to better convergence properties. A second improvement is a new technique for identifying the maximum achievable current density, an important piece of information necessary to avoid divergence of the implicit-equation solver. Third, a modeling improvement is introduced through the adoption of a linear ion-conductivity model that enhances the ability to better match experimental polarization-curve data at high current densities. Fourth, a systematic method is advanced for extracting anodic and cathodic transfer-coefficient parameters from experimental data via a least-squares regression procedure, eliminating a potentially significant parameter estimation error. Finally, this study determines that the methanol concentration boundary condition imposed on the membrane side of the membrane-cathode interface plays a critical role in the model’s ability to predict the limiting current density. Furthermore, the study argues for the need to carry out additional experimental work to identify more meaningful boundary concentration values realized by the cell.

SUPPORT VECTOR REGRESSION MODEL FOR DIRECT METHANOL FUEL CELL

2012 ◽  
Vol 23 (07) ◽  
pp. 1250055 ◽  
Author(s):  
J. L. TANG ◽  
C. Z. CAI ◽  
T. T. XIAO ◽  
S. J. HUANG
Keyword(s):  
Fuel Cell ◽  
Prediction Model ◽  
Support Vector Regression ◽  
Direct Methanol Fuel Cell ◽  
Cell Voltage ◽  
Percentage Error ◽  
Support Vector ◽  
Absolute Percentage Error ◽  
Direct Methanol ◽  
Methanol Fuel Cell

The purpose of this paper is to establish a direct methanol fuel cell (DMFC) prediction model by using the support vector regression (SVR) approach combined with particle swarm optimization (PSO) algorithm for its parameter selection. Two variables, cell temperature and cell current density were employed as input variables, cell voltage value of DMFC acted as output variable. Using leave-one-out cross-validation (LOOCV) test on 21 samples, the maximum absolute percentage error (APE) yields 5.66%, the mean absolute percentage error (MAPE) is only 0.93% and the correlation coefficient (R2) as high as 0.995. Compared with the result of artificial neural network (ANN) approach, it is shown that the modeling ability of SVR surpasses that of ANN. These suggest that SVR prediction model can be a good predictor to estimate the cell voltage for DMFC system.

Sign in / Sign up

Export Citation Format

Share Document