Gas Turbine Emissions Research Articles

Impacts of Environmental Conditions on Thermal, Emissions and Economic Performance of Gas Turbine Using Different Types of Fuels: An Experimental Investigation

Gas turbine has been used widely for electrical power generation based on simple cycle system. However, the relatively low thermal efficiency engages the researchers in investigating possible methods to enhance gas turbine performance. In the current study, three different types of fuels were used (crude oil, light oil, and natural gas) to operate the gas power plant in Qayyara, Mosul, Iraq and to evaluate the plant's performance, environmental impact and economic cost. The Aspen Husys package was used for simulation based on data recorded by the control department. Results showed that the software applied was highly valid. The external environmental temperature influenced the performance and efficiency of the station significantly with maximum values obtained for winter by 114MW and thermal efficiency of 32.02% for natural. The emissions of CO2, SO2, and NOX gases are decreased, whereas CO emission is increased when the temperature of the external environment is raised with fewer emissions obtained for natural gas. The cost of three fuels in (summer) is the highest among the four months, with the higher cost for light fuel by 68.2 Iraqi diners/kiloWatt.hour. Therefore, light fuel cannot be used for the permanent operation of gas turbines, and natural gas fuel gives better performance and lower emissions than the other two types of fuels.

Advanced hybrid neural network techniques for minimizing gas turbine emissions

Purpose Emissions have significant environmental impacts. Hence, minimizing emissions is essential. This study aims to use a hybrid neural network model to predict carbon monoxide (CO) and nitrogen oxide (NOx) emissions from gas turbines (GTs) to enhance emission prediction for GTs in predictive emissions monitoring systems (PEMS). Design/methodology/approach The hybrid model architecture combines convolutional neural networks (CNN) and bidirectional long-short-term memory (Bi-LSTM) networks called CNN-BiLSTM with modified extrinsic attention regression. Over five years, data from a GT power plant was uploaded to Google Colab, split into training and testing sets (80:20), and evaluated using test matrices. The model’s performance was benchmarked against state-of-the-art emissions prediction methodologies. Findings The model showed promising results for GT CO and NOx emissions. CO predictions had a slight underestimation bias of −0.01, with root mean-squared error (RMSE) of 0.064, mean absolute error (MAE) of 0.04 and R2 of 0.82. NOx predictions had an RMSE of 0.051, MAE of 0.036, R2 of 0.887 and a slight overestimation bias of +0.01. Research limitations/implications While the model demonstrates relative accuracy in CO emission predictions, there is potential for further improvement in future research. Practical implications Implementing the model in real-time PEMS and establishing a continuous feedback loop will ensure accuracy in real-world applications, enhance GT functioning and reduce emissions, fuel consumption and running costs. Social implications Accurate GT emissions predictions support stricter emission standards, promote sustainable development goals and ensure a healthier societal environment. Originality/value This paper presents a novel approach that integrates CNN and Bi-LSTM networks. It considers both spatial and temporal data to mitigate previous prediction shortcomings.

A Reactor-Network Framework to Model Performance and Emissions of a Longitudinally-Staged Combustion System for Carbon-Free Fuels

Abstract Hydrogen and ammonia are considered crucial carbon-free energy carriers optimally suited for seasonal chemical storage and energy system balance. In this context, longitudinally-staged combustion systems (LSC) represent an attractive technology for achieving low emissions while conserving high load and fuel flexibility at high efficiency. While these two-stage combustion systems have been successfully implemented for natural gas and hydrogen firing of gas turbines, optimal operation with ammonia-based fuel mixtures not yet established. Recent works have shown that ammonia can be combusted with low emissions in gas turbines by employing a Rich-Quench-Lean operational concept where a slightly rich fuel-air mixture is burnt in the first-stage followed by dilution air addition, which completely oxidizes the remaining unburnt fuel. However, an optimal operational strategy to efficiently combust all three fuels within the same system is yet to be established. In this work, we exploit a reactors-network framework to efficiently investigate the emissions performance of a LSC system fired with natural gas, hydrogen and ammonia. Firstly, the framework is validated with experimental and computational emission results in the literature. Secondly, the optimal operational strategy (in terms of fuel- and air-split between the stages) for clean and efficient ammonia-firing operation is found. Thirdly, the consequences of such ammonia-optimized operational strategy on flame stabilization and emissions with natural gas- and hydrogen are considered. Finally, the air and fuel distribution for optimal performance and minimal emissions for all three fuels is found for the LSC system.

Numerical Investigation of Effusion Cooling Air Influence On the Co Emissions for a Single-Sector Aero-Engine Model Combustor

Abstract Stricter aviation emissions regulations have led to the desire for lean-premixed-vaporized combustors over rich-quench-lean burners. While this operation mode is beneficial for reducing NOx and particulate emissions, the interaction of the flame and hot exhaust gases with the cooling flow results in increased CO emissions. Predicting CO in computational fluid dynamics (CFD) simulations remains challenging. To assess current model performance under practically relevant conditions, Large-Eddy Simulation (LES) of a lab-scale effusion cooling test rig is performed. Flamelet-based manifolds, in combination with the Artificial Thickened Flame (ATF) approach, are utilized to model the Turbulence-Chemistry Interaction (TCI) in the test rig with detailed chemical kinetics at reduced computational costs. Heat losses are considered via exhaust gas recirculation (EGR). Local transport effects in CO emissions are included through an additional transport equation. Additionally, a Conjugate Heat Transfer (CHT) simulation is performed for good estimations of the thermal boundary conditions. Extensive validation of this comprehensive model is conducted using the available experimental dataset for the studied configuration. Subsequently, model sensitivities for predicting CO are assessed, including the progress variable definition and the formulation of the CO source term in the corresponding transport equation. To investigate the flame thickening influence in the calculated CO, an ATF-postprocessing correction is further developed. Integrating multiple sophisticated pollutant submodels and evaluating their sensitivity offers insights for future investigations into modeling CO emissions in aero-engines and stationary gas turbines.

Hybrid Particle Swarm Optimization and Feedforward Neural Network Model for Enhanced Prediction of Gas Turbine Emissions

Hybrid Particle Swarm Optimization and Feedforward Neural Network Model for Enhanced Prediction of Gas Turbine Emissions

Effect of alternative biogas-methane fuels use on performance and emissions in turbojet engine

Researchers are trying to find alternative fuels to traditional fuels as traditional fuels are running out. This study is to predict turbojet gas turbine emission and performance using Methane and Biogas fuel blend for turbojet in Gasturb-14 program. Based on the physicochemical properties of each fuel, parameters were calculated for methane supplemented B1, B2 and B3 and pure methane fuel compositions. The heat of combustion for each blend was then obtained using GasTurb Details6 and the design points of the turbojet engine were determined. The program was run using biogas fuels to analyze performance and emission characteristics such as carbon monoxide, carbon dioxide, etc. In this study, the B3 blend was found to be the most effective Specific Fuel Consumption value with 32.22 g/kNs. It also produced lower emission rates for CO and CO2 compared to B1 and B2 fuels. The first finding is that it is possible to burn an alternative biogas fuel to methane fuel. The results also show that biogas is very similar to methane in terms of temperature distributions in the combustion chamber due to its high methane content compared to other biogases.

Experimental study on the performance of micro gas turbines under different intake environments

This article uses experimental methods to measure the operational performance and pollutant emissions of micro gas turbines in different intake environments, and analyzes the feasibility of replacing mine ventilation fans with micro gas turbines. The results show that when the compressor efficiency is increased from 70 % to 90 %, the total efficiency of the micro combustion engine is increased from 28% to 34 %, with an increase of up to 21.4 %. When the demand power is 25 kW, the fluctuation range of NOx emission concentration is within 5.4 %. When the intake opening is below 50 %, the effect is obvious. When the intake opening is reduced from 50 % to 25 %, the performance of the micro gas turbine decreases by 0.7 %; The influence of intake humidity on various operating parameters of micro gas turbines is generally within 3 %, which can meet the requirements of high humidity environment operation; The intake temperature has the greatest impact, with an increase in intake temperature leading to a decrease in power generation efficiency. After comprehensive analysis, we believe that the application of gas turbines in mines is feasible.

A Hybrid Neural Network‐Based Improved PSO Algorithm for Gas Turbine Emissions Prediction

AbstractIn gas‐fired power plants, emissions may reduce turbine blade rotation, thus decreasing power output. This study proposes a hybrid model integrating the Feed forward Neural Network (FFNN) model and Particle Swarm Optimization (PSO) algorithm to predict gas emissions from natural gas power plants. The FFNN predicts gas turbine nitrogen oxides (NOx) and carbon monoxide (CO) emissions, while the PSO optimizes FFNN weights, improving prediction accuracy. The PSO adopts a unique random number selection strategy, incorporating the K‐Nearest Neighbor (KNN) algorithm to reduce prediction errors. Neighbor Component Analysis (NCA) selects parameters most correlated with CO and NOx emissions. The hybrid model is constructed, trained, and testedusing publicly available datasets, evaluating performance with statistical metrics like Mean Square Error (MSE), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE). Results show significant improvement in FFNN training with the PSO algorithm, boosting CO and NOx prediction accuracy by 99.18% and 82.11%, respectively. The model achieves the lowest MSE, MAE, and RMSE values for CO and NOx emissions. Overall, the hybrid model achieves high prediction accuracy, particularly with optimized PSO parameter selection using seed random generators.

Life Cycle Assessment of a Gas Turbine Installation

Gas turbine installations (GTIs) are widely used to generate electrical and thermal energy, mainly by burning gaseous fuels. With the development of hydrogen energy technology, a current area of particular interest is the use of GTIs to burn hydrogen. In order to assess the prospects of using GTIs in this way, it is necessary to understand the carbon emissions of gas turbines within the larger context of the entire hydrogen life cycle and its carbon footprint. The article provides an overview of results from previously published studies on life cycle assessment (LCA) of complex technical devices associated with the production and consumption of fuel and energy, which are most similar to GTIs when it comes to the complexity of LCA. The subject of analysis was a set of GTIs located in Russia with a capacity of 16 MW. An assessment of greenhouse gas (GHG) emissions per MWh of electricity produced showed that at different stages of the GTI life cycle, the total carbon footprint was 198.1–604.3 kg CO2-eq., of which more than 99% came from GTI operation. Greenhouse gas emissions from the production and end-of-life management stages are significantly lower for GTIs compared to those for other complex technical devices used to generate electricity. This is an indicator of the strong prospects for the future use of GTIs.

Supervised learning-based multi-site lean blowout prediction for dry low emission gas turbine

Current dry low emission (DLE) gas turbines have extremely low nitrogen oxides (NOx) emissions, allowing them to comply with stringent environmental regulations. The ultra-low temperatures required to achieve such low emissions also increase the risk of a lean blowout (LBO). DLE gas turbines are particularly vulnerable to LBO, which can lead to system instability, higher carbon oxides (CO) emissions, damaged components, substantial financial losses and environmental damages. In this regard, this current study proposes a novel supervised learning-based prediction technique with efficient performance on out-of-distribution data from multiple DLE gas turbine plants (multi-site). The proposed prediction approach exploits the competitive advantages of both adaptive boosting (AdaBoost) and Linear support vector machine (LSVM) to improve the generalization capability as well as prediction accuracy. The proposed algorithm was trained and tested using a real-world DLE gas turbine dataset from six different sites. The result indicates that the proposed model consistently achieving LBO prediction accuracy rates above 99.9% and Mathew’s correlation coefficient (MCC) score above 0.9 across all datasets. All lean-premixed gas turbines could benefit from the developed Ada-LSVM, as it is an accurate model with a high generalization performance that can be used across multiple sites.

Study on MCFC-integrated GSCC systems with SEGR in parallel or series and CO2 capture

In this paper, two new molten carbonate fuel cell (MCFC)-integrated gas–steam combined cycle (GSCC) systems with selective exhaust gas recirculation (SEGR) and CO2 capture are proposed and analyzed. The CO2 concentration in the gas turbine emission is increased because CO2 is selectively recycled with the help of SEGR. Molten carbonate fuel cells (MCFCs) are another way to increase CO2 concentration in the gas turbine flue gas by translating only CO2 from the cathode to the anode. In these two new gas–steam combined cycle systems, SEGR connected with MCFC, either in parallel or series, increases CO2 concentration beyond 11%. A gas–steam combined cycle system combined with MCFC and CO2 capture without SEGR is used as the reference system. Aspen Plus software is adopted to build the system models, and the performances of different systems are discussed and compared. The research results reveal that for the MCFC-integrated gas–steam combined cycle system with SEGR in series and CO2 capture, the CO2 concentration of gas turbine exhaust increases to 11.72% and the thermal efficiency is 56.29% when the overall CO2 capture rate is 88.16%, which is 1.13% higher than that of the reference system; for the MCFC-integrated gas–steam combined cycle system with SEGR in parallel and CO2 capture, the CO2 concentration of gas turbine exhaust increases to 14.15% and the thermal efficiency is 56.62%, which is 1.46% higher than that of the reference system. Furthermore, the economic analysis results show that the economic performances of new systems are mainly influenced by MCFC cost and will be gradually improved with the decrease in the MCFC cost.

Tabular Machine Learning Methods for Predicting Gas Turbine Emissions

Predicting emissions for gas turbines is critical for monitoring harmful pollutants being released into the atmosphere. In this study, we evaluate the performance of machine learning models for predicting emissions for gas turbines. We compared an existing predictive emissions model, a first-principles-based Chemical Kinetics model, against two machine learning models we developed based on the Self-Attention and Intersample Attention Transformer (SAINT) and eXtreme Gradient Boosting (XGBoost), with the aim to demonstrate the improved predictive performance of nitrogen oxides (NOx) and carbon monoxide (CO) using machine learning techniques and determine whether XGBoost or a deep learning model performs the best on a specific real-life gas turbine dataset. Our analysis utilises a Siemens Energy gas turbine test bed tabular dataset to train and validate the machine learning models. Additionally, we explore the trade-off between incorporating more features to enhance the model complexity, and the resulting presence of increased missing values in the dataset.

Effect of Operating Conditions on the Emission Characteristics Of an Annular Combustor

Pollutant emissions from combustion processes have become of great public concern due to their impact on health and the environment. Recent growth in civil aviation and stationary gas turbines power plants is worsening the situation both at airport vicinity and in altitude. This has called for rapid changes both in the regulations and controlling gas turbine emissions and in the technologies used to meet these regulations. All these developments have been accompanied by continuous and increasing pressure on the combustion engineers to reduce pollutant emissions from all types of gas turbines while operating at very high levels of pressure and temperature. In this paper an effort is made to characterize an annular combustion chamber from emission point and effect of various operating conditions has been analyzed. The emission characteristics of combustor are very satisfied and are inline to that of contemporary combustors of same class. This study can lead to combustor modifications for bringing down the emission levels to meet the international regulations.

Prediction of Dry-Low Emission Gas Turbine Operating Range from Emission Concentration Using Semi-Supervised Learning.

Dry-Low Emission (DLE) technology significantly reduces the emissions from the gas turbine process by implementing the principle of lean pre-mixed combustion. The pre-mix ensures low nitrogen oxides (NOx) and carbon monoxide (CO) production by operating at a particular range using a tight control strategy. However, sudden disturbances and improper load planning may lead to frequent tripping due to frequency deviation and combustion instability. Therefore, this paper proposed a semi-supervised technique to predict the suitable operating range as a tripping prevention strategy and a guide for efficient load planning. The prediction technique is developed by hybridizing Extreme Gradient Boosting and K-Means algorithm using actual plant data. Based on the result, the proposed model can predict the combustion temperature, nitrogen oxides, and carbon monoxide concentration with an accuracy represented by R squared value of 0.9999, 0.9309, and 0.7109, which outperforms other algorithms such as decision tree, linear regression, support vector machine, and multilayer perceptron. Further, the model can identify DLE gas turbine operation regions and determine the optimum range the turbine can safely operate while maintaining lower emission production. The typical DLE gas turbine's operating range can operate safely is found at 744.68 °C -829.64 °C. The proposed technique can be used as a preventive maintenance strategy in many applications involving tight operating range control in mitigating tripping issues. Furthermore, the findings significantly contribute to power generation fields for better control strategies to ensure the reliable operation of DLE gas turbines.

Investigation of jet A-1 and waste cooking oil biodiesel fuel blend flame characteristics stabilized by radial swirler in lean pre-vaporized premixed combustor

Energy crisis and the new environmental legislations have motivated researchers worldwide to come up with new solutions to these emerging problems. One of the proposed solutions is to biodiesel from vegetable and waste oils to substitute the conventional fuels. The current study aims to investigate the flame characteristics of waste cooking oil biodiesel fuel blend stabilized by radial swirler in Lean Pre-vaporized Premixed (LPP) combustor. In this scope following tasks were performed (i) preparing Waste Cooking Oil Methyl (WCOME) via an ultrasonic assisted transesterification process, (ii) designing and constructing radial swirler (with swirl number SN = 0.55), and (iii) investigating the flame structures of base Jet A-1 fuel and a blend of WCOME and Jet A-1 containing 10% of WCOME (symbolized as B10). Experiments were performed at equivalence ratio Φ = 0.75 and air preheat temperature of 310 °C. Measurements revealed more temperature distribution with lower maximum value for B10 than those for base fuel. CO was higher for B10 at exit level in the range of 14,000 ppm. On the other hand, it was on average of 230 ppm for Jet A-1. NOx emissions recorded was slightly lower for B10 than Jet A-1, but both were about 10 ppm. It could be concluded that, B10 can replace Jet A-1 to reduce NOx emissions with uniform temperature distribution. Use of LPP combustion reduces gas turbine emissions and use of radial swirler to stabilize the lean premixed mixture is confirmed.

Automatized Experimental Combustor Development Using Adaptive Surrogate Model-Based Optimization

Abstract Lean premixed combustion is the state-of-the-art technology to achieve ultra low NOx emissions in stationary gas turbines. However, lean premixed flames are susceptible to thermoacoustic instabilities, lean blowout, and flashback. The design of such a combustion system is thus always related to the balancing between the levels of emissions and flame stability. Data-driven optimization methods and the adaptation of models through artificial intelligence have experienced a surge in development in the past years. The goal of this study is to show the potential of these methods for gas turbine burner development. A special pilot burner that features 61 different positions of fuel injection, manufactured by means of selective laser melting is used to modify the gas mixture close to the flame anchoring position. Each of the injector lines is equipped with an individual valve, such that the distribution of fuel-air mixture can be modified variously. Installed into an industrial MGT6000 swirl combustor, a data-driven optimization method is used to find an optimal subset of injection locations by automated experiments. The method uses a surrogate model that is based on Gaussian processes regression. It is adopted for experimental optimization, keeping measurement efforts to a minimum. The optimizer controls the fuel valves and uses live measurements to find a distribution that generates minimal NOx emissions while ensuring flame stability. The solutions found by the optimization scheme are analyzed and advantages and limitations of the approach are discussed.

An Optimization Methodology for Turbines Driven by Pulsed Detonation Combustors

Abstract A step-change in efficiency of gas turbine technology and, subsequently, an emissions reduction from this technology requires conceptual changes. Substituting conventional combustion chambers with pressure gain combustion in the form of pulsed detonation combustion (PDC) is one of the promising methods that can reduce gas turbine emissions significantly. Nevertheless, the component matching for the respective systems and specifically that of turbine expanders working with the exhaust flow of PDC tubes is still not solved. The unsteady nature of PDC exhaust flow makes 3D-CFD simulations too expensive to be applied in optimization loops in early design stages. To address this question, the present paper introduces a new cost-effective but reliable methodology for turbine analysis and optimization, based on the unsteady exhaust flow of pulsed detonation combustors. The methodology unitizes a robust unsteady one-dimensional solver, a meanline performance analysis, and an adaptive surrogate optimization algorithm. A two-stage axial turbine is optimized considering all unsteady flow features of a hydrogen-air PDC configuration with five PDC tubes. A three-dimensional URANS simulation is performed for the optimized geometry and the baseline to evaluate the methodology. The results showed that the optimized turbine produces 16% lower entropy than the original one. Additionally, the turbine output power is increased by 14% by the optimized design. Based on the results, it is concluded that the approach is fast and reliable enough to be applied in optimizing any turbine working with unsteady flows, more specifically in PDC applications.

Evaluation of a Turbulent Jet Flame Flashback Correlation Applied to Annular Flow Configurations With and Without Swirl

Abstract The adaptation of high hydrogen content fuels for low emissions gas turbines represents a potential opportunity to reduce the carbon footprint of these devices. The high flame speed of hydrogen air mixtures combined with the small quenching distances poses a challenge for using these fuels in situations where a significant premixing is desired. In particular, flashback in either the core flow or along the walls (i.e., boundary layer flashback) can be exacerbated with high hydrogen content fuels. In this work, the ability of a flashback correlation previously developed for round jet flames is evaluated for its ability to predict flashback in an annular flow. As a first step, an annular flow is generated with a centerbody located at the centerline of the original round jet flame. Next, various levels of axial swirl is added to that annular flow. Additional flashback data are obtained for various mixtures of hydrogen and methane and hydrogen and carbon monoxide for all the annular flow configurations. Pressures from 3 to 8 bar are tested with mixture temperatures up to 750 K. Flashback is induced by slowly increasing the equivalence ratio. The results obtained show that the same form of the correlation developed for round jet flames can be used to correlate flashback behavior for the annular flow case with and without swirl despite the presence of the centerbody. Adjustments to some of the constants in the original model were made to obtain the best fit, but in general, the correlation is quite similar to that developed for the round jet flame. A significant difference with the annular flow configurations is the determination of the appropriate gradient at the wall, which in the present case is determined using a cold flow computational fluid dynamics simulation.

Effects of adiabatic flame temperature on flames’ characteristics in a gas-turbine combustor

In this study a comparison between the premixed methane oxygen-enriched-air (CH4/O2/N2) and oxy-methane (CH4/O2/CO2) in a gas-turbine model combustor that imitates pre-mixers in operational air-fuel dry low emissions gas turbines is presented. The comparison and analyses that follow in this study are based on the results of experiments conducted. The combustion stability maps were obtained through the estimation of acoustic limits and measurement of blowout limits within the space of equivalence ratio (φ) – oxygen fraction (OF). The stability maps obtained were superimposed on the contour plots of constant adiabatic temperature (Tad), Reynolds number (Re), and power density (PD) of the combustor. Effects of Tad on flame macrostructure, flame stability, flame speed, and blowout mechanism were investigated. The temperature distributions were also measured. The study results indicated that blowout of CO2 as well as N2 flames occur at constant Tad. These results are more pronounced in the case of the multi-hole burner because, such burner is not characterized with sporadic nature of flame lifting and reattachment that dominate the occurrence of a blowout in swirl burner; for a given OF, φ at which CO2 flames blowout is higher than that of N2 flames due to the poorer resistance of oxy-flames to blowout as compared to air flames; the stable combustion zone of CO2 flames is larger than that of N2 flames.

The Impacts of Degradation on Gas Turbine Emission Evaluation in Natural Gas Pipeline Application

The Impacts of Degradation on Gas Turbine Emission Evaluation in Natural Gas Pipeline Application