- : Ms Word, Ms Word Format
- : 100 Pages
- : ₦5000
- : 1-5 Chapters
- Click to DOWNLOAD Materials
EXPERIMENTAL STUDY AND ARTIFICIAL NEURAL NETWORK PREDICTION OF THE PERFORMANCE AND EMISSION CHARACTERISTICS OF SPARK IGNITION ENGINE RUNNING ON ETHANOL/PETROL BLENDS
ABSTRACT
In this study, the performance and exhaust emission analysis of a single cylinder spark ignition engine fuelled with extended range of ethanol – petrol blends were carried out at constant load and constant engine speed conditions. Ethanol produced from Nigerian feedstock was blended with petrol at different proportions by volume. In order to establish a baseline for comparison, the engine was first run on neat petrol. The engine performance and exhaust emission parameters were determined for each blend of fuel at different engine load and speed. Also, in this study a Levenberg Marquardt Artificial Neural Network (ANN) algorithm and Logistic sigmoid activation type transfer function with a 4–15–3 model was developed to predict the brake specific fuel consumption (BSFC), combustion efficiency, maximum pressure, the CO and CO2 emissions of G200 IMEX spark ignition engine using the recorded engine speed, engine torque, bioethanol mixtures, mass flow rate and exhaust gas temperature as input variables. The performance of the ANN was validated by comparing the predicted data with the experimental results. The experimental results showed that addition of ethanol to petrol causes an improvement in engine performance and significant reduction in CO and HC emissions. At 2500 rpm, E30 recorded the maximum thermal efficiency of 43.20% followed by E20 with thermal efficiency of 42.8% while the maximum thermal efficiency for petrol fuel at 2500 rpm was 24.12%. E30, E20, E15 and E10 blends showed respectively an increment of 7.25%, 14.50%, 10.14% and 4.35% in combustion efficiency compared to that of petrol. Generally, the CO and HC emissions were significantly reduced through blending of petrol with ethanol. In all, ethanol and its blends with petrol exhibited performance characteristics trends similar to that of petrol thus suggesting them as suitable alternative fuels for spark ignition engines. The results from the model showed that the training algorithm was sufficient enough to predict the performance of the test engine. Correlation coefficient (R) of 0.974, 0.996, 0.995, 0.9479 and 0.939 were obtained for brake specific fuel consumption, combustion efficiency, maximum pressure, carbon dioxide emission and carbon monoxide emission respectively. These correlation coefficient (R) obtained for the output parameters are very close to 1 showing good correlation between the ANN predicted results and the experimental data while the Mean Square Error (MSE) were found to be very low (0.00018825 @ epoch 10 for BSFC, 1.0023 @ epoch 3 for combustion efficiency, 0.0013284@ epoch 5 for in-cylinder pressure, 0.00021234@ epoch 5 for CO2 and 0.0022503@ epoch 4 for CO emission). The results suggest that ANN can be a useful tool for the prediction of performance and emission characteristics of spark ignition engines under varying operating conditions. This will help to reduce both cost and time required for experimental studies of spark ignition engines.
Keywords: Artificial Neural Network; Performance and Emission; Ethanol – Petrol Blends; Combustion Efficiency
TABLE OF CONTENTS
Title Page i
Certification ii Dedication iii Acknowledgement iv Table of Content v List of Tables vii List of Figures viii Nomenclature x Abstract xii
CHAPTER ONE INTRODUCTION
1.1 Background Information 1
1.2 Problem Statement 5
1.3 Objectives of the Work 6
1.4 Justification of Study 6
1.5 Scope of Study 8
CHAPTER TWO LITERATURE REVIEW
2.1 Fuels 9
2.1.1 Solid Fuels 10
2.1.2 Petroleum (Liquid) Fuels 11
2.1.3 Properties of Hydrocarbon Fuels 19
2.1.4 Gaseous Fuels 24
2.1.5 Synthetic Fuels 26
2.1.6 Biofuels – Bioenergy and Biofuels an Overview 29
2.1.7 Biodiesel 34
2.1.8 Bioethanol 42
2.2 Engine Performance Parameters 67
2.2.1 Engine Power 68
2.2.2 Engine Torque 69
2.2.3 Specific Fuel Consumption 70
2.2.4 Friction Power 71
2.2.5 Mean Effective Pressure 71
2.2.6 Mechanical Efficiency 72
2.2.7 Thermal Efficiency 73
2.2.8 Volumetric Efficiency 74 2.2.9
Combustion Efficiency 74
2.3 Neural Network 79
2.3.1 Neuron 79
2.3.2 Artificial Neural Networks 80
2.3.3 Architecture 82
2.3.4 Number of Nodes and Layers 83
2.3.5 Setting Weights 83
2.3.6 Running Neural Network and Activation Function 83
2.3.7 The Back Propagation Algorithm 84 2.3.8 Review of Previous Works on Ann 88
2.4 Research Gap 89
CHAPTER THREE MATERIALS AND METHODS
3.1 Materials 91
3.1.1 Fuel Samples 91
3.1.2 The KM9106 Exhaust Gas Analyser 92
3.1.3 Engine Test Rig and Other Instruments 94
3.2 Methods 94
3.2.1 Experimental Set Up 94
3.2.2 Test Procedures 96
3.2.2.1 Constant Load Test 96 3.2.2.2 Constant Speed Test 97
3.2.2.3 Internal Combustion Engines Performance Equations 97
3.2.3 Neural Network Design 99
3.2.3.1 Artificial Neural Networks (ANNs) Structure 99
3.2.3.2 Artificial Neural Network Model and Parameters 103
CHAPTER FOUR RESULTS AND DISCUSSION
4.1 Results 105
4.1.1 Experimental Results 105 4.1.1.1 Constant Load Test 105 4.1.1.2 Constant Speed Test 113
4.1.2 Artificial Neural Networks Results 120
4.2 Discussion 131
4.2.1 Discussion of Experimental Results 131
| 4.2.2 Discussion of Artificial Neural Networks Results | 138 | |||
| CHAPTER FIVE
CONCLUSION AND RECOMMENDATIONS |
||||
| 5.1 Conclusion | 141 | |||
| 5.2 Contributions to Knowledge | 143 | |||
| 5.3 Recommendations | 143 | |||
| REFERENCES | 144 | |||
| APPENDIX A | 154 | |||
| APPENDIX B | 156 | |||
CHAPTER ONE INTRODUCTION
1.1 Background Information
The ever rising cost of fossil fuel internationally has forced major world economies, which are also major importers of fossil fuel, to resort to renewable and cheaper alternatives to fossil fuel to meet their energy demands. The limited nature of oil resources has made studies on alternative energy sources much more important in internal combustion engines in which oil products are used as an energy source
[Canakci et al., 2006; Hulwan and Joshi, 2011; Kannan et al., 2011; Fahd et al.,2013]. Biodiesel and bio-ethanol have emerged as the most suitable renewable alternatives to fossil fuel as their quality constituents match diesel and petrol respectively. In addition they are less polluting than their fossil fuel counterparts. Environmental concerns and the desire to be less dependent on imported fossil fuel have intensified worldwide efforts for production of biodiesel from vegetable oils and ethanol from starch and sugar producing crops [Igbokwe, 2012] .
The petrol and diesel engines are one of the most efficient power plants in use today; consequently, they enjoy wide application in road, rail and marine transportation as well as power generation. Petrol and diesel are fossil fuel and whose production and combustion result in the emission of gases that have adverse effects on human health and environment. The greenhouse gas emissions from the combustion of hydrocarbon fuels have been identified as the major causes of climate change and global warming. The numerous and varied effects of climate change on the environment, human life and the economy of the nations are becoming increasingly obvious and real. Expectedly, the phenomenon is provoking global concern as shown by world summits, local and international conferences on climate change and global warming held at different places and times like the Kyoto, Japan in 1997, Hague, Netherlands in 2000 and Copenhagen, Denmark in December 2009. Pragmatic steps like setting of drastic emission cut targets for both developing and developed nations have been taken. Global warming is the increase in the average temperature of Earth‘s near surface air and oceans. This increase in global temperature has caused changes in the global climate patterns described as climate change [Ofoh, 2009; Lu, et al., 2007].
Both the Intergovernmental Panel on Climate Change (IPCC) Report of 2007 and the U S Energy Departmental report of 2008 associated global warming with rise in sea level, flooding, changes in rainfall pattern, deforestation, glacier retreat, increased precipitation and high hurricane power dissipation index. On human health, global warming has been linked with increase in cardiovascular diseases, asthma, and other lung diseases due to the concentration of ozone at ground level [McMichael, et al., 2003]. Hales, et al., [2002] , and Rogers,et al., [2004] associated wide spread of diseases like dengue fever and malaria with global warming. There are predictions that global warming may substantially fuel itself by causing the loss of carbon from terrestrial ecosystem leading to an increase in the level of atmospheric carbon dioxide and the release of methane gas from methane clathrate [Cox, et al., 2007]. Both carbon dioxide and methane are greenhouse gases.
The emerging consensus is that global warming in the recent decades has been caused primarily by the greenhouse gas emissions from the production and processing of fossil fuels, their combustion in the factories, power generation, transportation and other human activities like deforestation and urbanization [Ofoh, 2009; Barma, et. al., 2010]. These activities have increased the concentration of greenhouse gases in the atmosphere. The concentrations of carbon dioxide and methane have increased by 36% and 148% respectively and fossil fuel burning has produced three-quarters of the increase in carbon dioxide from human activity over 20 years [EPA, 2007; IPCC, 2000; Pearson, and Palmer, 2000] . Transportation in industrialized and developing countries constitutes the largest and most rapidly growing user of world‘s oil reserve and hence a major source of greenhouse emissions [Giri, 2004,] .
Alternative fuels from bio-resources are considered biodegradable, renewable and environmentally friendly [Prasad, et al., 2010]. Among the alternative fuels that are gaining global interest particularly for internal combustion engines are bio-fuels like Bioethanol and Biodiesel. While bio-ethanol is considered a good alternative fuel for petrol engines, biodiesel is considered as good alternative for diesel engines.
In this study, experimental evaluation of the performance and emission characteristics of spark ignition engine running on different ethanol petrol blends at constant load and constant speed conditions was undertaken. It is intended to generate pertinent operating parameters for the blends with a view to identifying the blend ratio for optimal performance and emission. Because conducting such performance experiments on engines at different operating conditions require energy, money and time, artificial neural network (ANN) can be developed and used in order to decrease cost and save time [Ghazikhani and Mirzali, 2011] . ANN is a Meta heuristic artificial intelligent tool that has the ability to generalize between the input data and the target to give an output. It also has the capability to relearn and adapt for improving its performance with the availability of updated data. Once the ANN model is properly and sufficiently trained, it can generalize to similar cases [Sharkey et al., 2000; Yilmaz and Bilgin, 2013]. Different variables can be predicted by using the data obtained in previously conducted experiments. In recent years the applicability of an artificial neural network method for internal combustion engines has gained considerable success [Sharkey et al., 2000; Yilmaz and Bilgin, 2013] but skewed to diesel engines.
The existing literature has shown that ANN is a powerful modeling tool that has the ability to identify complex relationships between input and output data. Therefore, this work is geared towards developing a neural network model to predict the performance of a single cylinder spark ignition engine in relation to input variables including engine speeds, engine torque, mass flow rate, ethanol-petrol mixtures and exhaust gas temperature. This study is divided into: the experimental evaluation of the performance and emission characteristics of an extended range of ethanol petrol blends different operating conditions; determination of the effect of blends on the performance parameters obtained; the application of artificial neural network for predicting some of these performance and emission parameters mentioned above and then validating the network by comparing the result with the experimental data.
This model is of a great significance due to its ability to predict engine performance under different conditions [Banapumath et al., 2008]. Experiments were carried out using petrol-ethanol blends of different proportions at different engine speed in a single cylinder four stroke G200 IMEX engine coupled to a brake dynamometer. The exhaust of the engine is connected to a KM9106 exhaust gas analyzer. A feed forward back propagation algorithm was used for the ANN structure.
1.2 Problem Statement
Humanity has depended on fossil fuels for many decades as its major source of energy. This has led to increase in the demand and use of fossil fuels with some far reaching economic and environmental consequences. The increase in the demand of fossil fuels has led to incessant rise in their prices as well as the fear of exhaustion of their reserves. These challenges have provoked global search for alternative fuels that are renewable, cost effective, and environmentally friendly. Biodiesel and bio-ethanol are widely speculated to be the possible candidate fuels to meet these needs due to their low emission profiles. The production of biofuels is currently engaging the attention of the international community. Nigeria with large population and high consumption rate of fossil fuels is expected to be actively involved in the development of bio-fuels in order to cut down greenhouse emissions and make up for the shortfall in local fuel supply.
Conducting performance experiments on engines under different operating conditions using different fuels require cost and time. At this point, artificial neural network (ANN) can be used to reduce cost and save time required for the rigorous experimental process. In recent years the applicability of an artificial neural network method for internal combustion engines has gained considerable success [Sharkey et al., 2000; Yilmaz and Bilgin, 2013] but skewed to diesel engines. The existing literature has shown a wide gap in the application ANN in the prediction of the performance and emission parameters of SI engines considering the nature of fuel, the range of fuel blends used and the number of parameters so far predicted.
This project represents an effort to address the earlier identified problems associated with the development of alternative fuels from organic and renewable sources (such as high cost of biofuels production process) which will have performance characteristics similar to conventional (fossil) fuel. This will not only contribute significantly towards domestication of this technology but also enlist Nigeria among the nations making frantic efforts to cut down greenhouse emissions in order to mitigate the effects of climate change.
1.3 Objectives of Study
The primary objective of this research work is the experimental study and artificial neural network prediction of the performance and emission characteristics of a spark ignition engine running on ethanol-petrol blends. The specific objectives include:
- To measure the relevant engine parameters for different ethanol/petrol blends at constant load and constant speed conditions;
- To use the above data and the appropriate engine performance equations to calculate the engine performance;
- To measure the exhaust gas emission using the exhaust gas analyzer;
- Development of Artificial Neural Network model for prediction of the performance emission parameters of the test engine.
- Validation of the neural network performance by comparison with this experimental data.
1.4 Justification of Study
This research work is justified by the current global search for alternative fuels and energy sources that are both renewable and environmentally friendly. This search has been informed by the obvious negative consequences of the over dependence of humanity on fossil fuels for transportation, industry, power generation etc. These fossil fuels are exhaustible and this raises the issue of energy security. They are also the major sources of greenhouse gases like carbon dioxide which is a major contributor to global warming and climate change. This work involves the development of fuel that is generally considered as cleaner, biodegradable and non-toxic. This is believed to contribute to the global efforts to control the environmental effects of the use of fossil fuels such as global warming.
This work is also justified by the need to develop good artificial intelligent systems that can help reduce the time, energy and cost requirements of biofuels production process. It is crucial that all aspects of the bio-energy production process are streamlined and improved, from the design of more efficient bio-refineries to research into biofuels as an energy carrier. Current energy infrastructures need to be adapted and changed to fulfill the promises of biomass for power generation. As biofuels research continues at an unprecedented rate, the development of new feedstocks and improvements in bioenergy production processes provide the key to the transformation of biomass into global energy resources.
This work will stimulate interest in the harnessing of these abundant bio-energy resources available in the country. This has the multiplying effects of promoting agricultural activities, generation of rural employment, provision of rural infrastructures and the enhancement of the standard of living of rural dwellers. Nigeria is endowed with huge natural resources and factors deployable for bio-energy production like large arable land and favourable climate conditions.
1.5 Scope of Study
This work is focused on the experimental study and performance prediction of spark ignition engine running on ethanol-petrol blends using artificial neural network. It covers the following areas:
Performance experiments on the test engine at different operating conditions using different blends of ethanol with petrol; evaluation of the effect of blends on some of the vital performance parameters obtained; the application of artificial neural network for predicting some of these performance parameters mentioned above and then validating the network by comparison with the experimental data.
EXPERIMENTAL STUDY AND ARTIFICIAL NEURAL NETWORK PREDICTION OF THE PERFORMANCE AND EMISSION CHARACTERISTICS OF SPARK IGNITION ENGINE RUNNING ON ETHANOL/PETROL BLENDS
