NOx Emission Model Research Articles

Predictive Modeling of NOx Emissions from Lean Direct Injection of Hydrogen and Hydrogen/Natural Gas Blends Using Flame Imaging and Machine Learning

Predictive Modeling of NOx Emissions from Lean Direct Injection of Hydrogen and Hydrogen/Natural Gas Blends Using Flame Imaging and Machine Learning

Vehicle emission models alone are not sufficient to understand full impact of change in traffic signal timings

Few studies have considered the real-world impact of changes in traffic signal timings on air pollution and pedestrian exposure with most only drawing their conclusion from vehicle emission models alone. Here, we consider two distinct cycle timings at a junction in London, UK, model the impact using a traffic microsimulation and a NOx emissions model, and compare these results with NOx and other air pollution measurements collected during a two-week field study at the junction.Our models predict that extending the cycle time leads to a 23% decrease in NOx emissions within a 15 m radius of the junction itself. When the wind direction was such that our sensors were downwind of the junction a 21% decrease in traffic and background-adjusted NOx concentrations were seen, suggesting that the intervention was successful. However, when the sensors were upwind of the junction, we observed an increase of 23% in adjusted NOx concentrations. Similar patterns were found for the other pollutants NO2, lung deposited surface area, black carbon and CO2 we measured. This indicates that meteorology was by far the greatest determinant of roadside concentrations during our two-week study period.Looking at pedestrian exposure for pedestrians waiting to cross the road, we found that their NOx exposure increased by 46% as waiting times to cross the road increased and that potential small reductions in air pollution were offset by increases in waiting times on the main road.The study demonstrates the need to go beyond assessing the impact of hyper-local traffic interventions on vehicle emissions. Real-world trials over extended periods are required to evaluate the impact of meteorology and changes to air pollution concentrations and pedestrian exposures.

Dispersion model of NOx emissions from a liquefied natural gas facility

Natural gas used widely in terms of energy production. Energy production is among the most prominent sectors of humankind. Combustion processes inevitably produces air pollutants. The major pollutant during a combustion process is nitrogen oxide emissions. The term of nitrogen oxides primarily include nitrogen monoxide and nitrogen dioxide. These pollutants are generated regardless of the fuel content since air composition itself is the major source for these pollutants. It is possible to calculate emissions through the activity data and emission factors. Calculation of emissions is not enough for an environmental assessment. The impact of pollutants on human health relies on their concentration in the atmosphere. In order to determine their concentrations several modelling practices are developed. In this study, AERMOD used for modelling purpose of NOx emissions from a liquefied natural gas facility. It was observed that the pollutants were dispersed mostly towards south-southwest of the facility, where Marmaraereğlisi district is located. Although the pollutants transported directly to the settlement, the concentrations remained limited. During operation conditions, the highest daily NOx concentration was 1.7 μg/m3 and the highest annual concentration was 0.1 μg/m3. At maximum operating conditions, the highest daily NOx concentration was 16.2 μg/m3 and the highest annual concentration was 2.5 μg/m3. At minimum operating conditions, the highest daily NOx concentration was 1.1 μg/m3 and the highest annual concentration was 0.2 μg/m3.

Soft computing-based modelling and optimization of NOx emission from a variable compression ratio diesel engine

Machine learning method and statistical method used for model prediction and optimization of third generation biodiesel-diesel blend powered variable compression engine High R2 values of 0.9998 and 0.9994 were observed in the training and testing phase of the model, respectively, indicating that The results confirm the robustness of the forecasting system. It was shown that the model accuracy means squared errors remained low at 0.0002 and 0.0014. These results were then confirmed by desirability-based optimization, which succeeded in achieving the values of the set parameters It should be noted that the compression ratio (CR), fuel injection pressure, and engine load were optimized to meet the defined parameters, resulting in a NOx emissions reduction as 222.8 ppm. The research illustrates the efficacy of desirability-based optimization in attaining targeted performance targets across important engine parameters whilst also reducing the impact on the environment.

CPFD modeling of hydrodynamics, combustion and NOx emissions in an industrial CFB boiler

The ultra-low NOx emission requirement (50 mg/m3) brings great challenge to CFB boilers in China. To further tap the NOx abatement potential, full understanding the fundamentals behind CFB boilers is needed. To achieve this, a comprehensive CPFD model is established and verified; gas-solid flow, combustion, and NOx emission behavior in an industrial CFB boiler are elaborated; influences of primary air volume and coal particle size on furnace performance are evaluated. Simulation results indicate that there exists a typical core-annular flow structure in the boiler furnace. Furnace temperature is highest in the bottom dense-phase zone (about 950 °C) and decreases gradually along the furnace height. Oxygen-deficient combustion results in high CO concentration and strong reducing atmosphere in the lower furnace. NOx concentration gradually increases in the bottom furnace, reaches maximum at the elevation of secondary air inlet, and then decreases slightly in the upper furnace. Appropriate decreasing the primary air volume and coal particle size would increase the CO concentration and intensify the in-furnace reducing atmosphere, which favors for NOx reduction and low NOx emission from CFB boilers.

Dynamic Modeling of Nox Emission from Power Plants Considering Input Variable Time-Delay and Feature Extraction

Aiming at the problem that it is difficult to effectively predict the concentration of nitrogen oxide emissions from boilers under the changing load conditions of power plants, this paper proposes a dynamic modeling method of nitrogen oxide (NOx) emissions from power plants, which considers the delay-feature extraction of input variables. This method combines principal component analysis and Gaussian regression modeling, and uses principal component analysis to extract the feature information extracted from the input variable data, and the current value and historical sequence value of extracted information are used as the input of Gaussian regression model. In addition, the historical time series value of boiler NOx concentration is added to the input of the model as feedback data. Taking the combustion system of a 1,000MW ultra-supercritical unit boiler as an object, and combining with the actual operation data on site, a dynamic model of boiler NOx emission is established. The experimental results show that the built boiler NOx emission dynamic model has high prediction accuracy and strong generalization performance, which has a certain reference value for boiler NOx emission modeling and intelligent control research.

Long-Term Prediction Model for NOx Emission Based on LSTM–Transformer

Excessive nitrogen oxide (NOx) emissions result in growing environmental problems and increasingly stringent emission standards. This requires a precise control for NOx emissions. A prerequisite for precise control is accurate NOx emission detection. However, the NOx measurement sensors currently in use have serious lag problems in measurement due to the harsh operating environment and other problems. To address this issue, we need to make long-term prediction for NOx emissions. In this paper, we propose a long-term prediction model based on LSTM–Transformer. First, the model uses self-attention to capture long-term trend. Second, long short-term memory network (LSTM) is used to capture short-term trends and as secondary position encoding to provide positional information. We construct them using a parallel structure. In long-term prediction, experimental results on two real datasets with different sampling intervals show that the proposed prediction model performs better than the currently popular methods, with 28.2% and 19.1% relative average improvements on the two datasets, respectively.

Prediction model of NOx emissions in the heavy-duty gas turbine combustor based on MILD combustion

A prediction model of NOx emissions in heavy-duty gas turbine combustors based on the moderate and intense low-oxygen dilution (MILD) combustion was proposed, and the inlet parameter sensitivity analysis was also investigated to identify the impact weight for the practical operation of heavy-duty gas turbines. The optimal space-filling (OSF) design was first adopted to determine the optimal combination of gas temperature (Tgas), primary air temperature (Tfirst), secondary air temperature (Tsecondary), gas mass flow (mgas), primary air mass flow (mfirst), and secondary air mass flow (msecondary) for minimum NOx emissions. Based on the results of the OSF design, the minimum NOx emissions of 24.11 mg/m3 were obtained, and the corresponding Tgas, Tfirst, Tsecondary, mgas, mfirst, and msecondary were 769.67 K, 825.10 K, 722.61 K, 1.71 kg/s, 4.29 kg/s, and 7.69 kg/s within the range of 10% fluctuation of rated input parameters. The sensitivity ranking of the six inlet parameters was secondly achieved as mfirst > Tfirst > Tsecondary > msecondary > mgas > Tgas with the base of NOx emissions. Finally, a novel prediction model with six inlet parameters was established to quickly and accurately calculate the NOx emissions influenced by complicated boundary conditions in MILD combustion of heavy-duty gas turbines.

Kinetics modeling of NOx emission of oxygen-enriched and rich-lean-staged ammonia combustion under gas turbine conditions

Though the oxygen enrichment combustion and fuel-rich/quick-mix/fuel-lean (RQL) combustion are respectively effective to improve burning velocity and reduce NOx emissions respectively, their combination feasibility and characteristics for burning ammonia (NH3) need further investigation as oxygen enrichment could increase NOx production. Thus, a kinetic modeling study was conducted on the NOx emission characteristics of an oxygen-enriched and RQL-staged NH3-fired combustor under gas turbine conditions by using the chemical reaction network (CRN) method. The results revealed that NOx emission could be minimized as long as the equivalence ratio (ϕ) in the fuel-rich stage was kept at a certain value according to the oxygen enrichment degree (Ω) in the fuel-rich stage oxidizer, regardless of the Ω in the fuel-lean stage oxidizer. At a higher Ω, low NOx emission was achieved in a wider ϕ range, while the optimal ϕ in the fuel-rich stage became higher and the corresponding NOx emission was lower. This mainly benefited from the stronger NH3 consumption in the fuel-rich stage due to the higher temperature and higher concentration of OH/H radicals. However, a higher Ω in the fuel-rich stage enhanced the thermal NOx formation in the fuel-lean stage. Based on the results of the present study, oxygen-enriched NH3 combustion with a proper ϕ setting in the fuel-rich stage could simultaneously improve the burning intensity and reduce NOx emission for burning NH3 in a gas turbine.

Development of a NOx Calculation Model for Low-Speed Marine Diesel Engines Based on Soft Measurement Technology

With the increasing level of intelligence of marine engines, there is an increasing demand for the online monitoring of engines, and marine NOx emissions have been of great concern. In this paper, a NOx simulation model is developed based on virtual measurement technology, which can calculate and predict NOx emissions based on the current operating state parameters of low-speed two-stroke diesel engines. First, the calibrated 3D simulation model is used to design the experiments to obtain the simulation experimental samples. Based on the NOx generation mechanism and diesel engine work-related parameters, the relevant factors were selected as alternative input parameters for the NOx emission model. The correlation analysis was then performed on the input parameters using the grey relational analysis correlation method and the Pearson correlation coefficient, and the principal component analysis method was used to reduce the dimensionality of the relevant factors by minimizing the loss of important information in reducing the complexity of the whole model. Then, the structure-related parameters of the backpropagation neural network (BPNN) were adaptively optimized using the group method of data handling (GMDH) to improve the accuracy of the NOx soft measurement model. Finally, the developed GMDH–BP model was validated with data and compared with the error evaluation index of BPNN and BPNN optimized by genetic algorithm (GA), and the developed NOx simulation model demonstrated high prediction accuracy under the same hyperparameter settings. The result provides technical support for the subsequent realization of the real-time online monitoring of NOx emissions from low-speed marine diesel engines without NOx sensors.

Optimized fluegas denitrification control strategy to enhance SCR performance during load-cycling transient processes in coal-fired power plants

Selective catalytic reduction (SCR) is a common method for controlling NOx emission via NH3 injection. Coal-fired power plants are taking load-cycling processes with increasing frequency. During load-cycling processes, NOx emissions fluctuate and NH3 escape aggravates, which increases environmental pollution and degrades equipment safety. Therefore, the NH3 injection control strategy must be optimized to enhance denitrification performance during transient processes. In this study, an integrated NOx emission model of coal-fired units is developed and verified. The model includes the thermal system simulation, NOx production, and SCR system models. On the basis of the model, load-cycling processes are simulated under the traditional denitrification control strategy, wherein excessive NH3 injection and increased NOx and NH3 emissions are observed. The typical transient processes of the SCR system and boiler are simulated with various initial operating conditions and input parameter disturbances to explore the main factors affecting denitrification performance during load-cycling processes. The analysis of dynamic characteristics reveals that the inertia of NH3 storage is an important factor of unstable denitrification performance, and the temperature and fluegas flow rate influence the delay caused by this inertia, therefore the denitrification performance shows significant difference in various load-cycling processes. Consequently, an optimized NH3 control strategy is proposed. In this strategy, the change rate of NOx concentration is predicted according to the boiler load command, and the fluegas temperature and flow rate are also taken to adjust the additional NH3 injection. The simulation of typical load-cycling processes demonstrates that the optimized control strategy comprehensively enhances denitrification performance, where the NH3 injection, NOx emission and NH3 emission are all reduced. In particular, for the low- and medium-load range, the total amount of NH3 emission can be reduced by 5.69%, and the maximum values of NOx and NH3 emissions are reduced by 7.46% and 19.26%, respectively, which benefits both air pollutants emission reduction and equipment safety.

NOx emissions by real-world fresh and old asphalt mixtures: Impact of temperature, relative humidity, and UV-irradiation

Air pollution causes chronic illnesses, and an estimated 4.5 million deaths every year. 70% of mankind is expected to live in cities by 2050, making urban areas key for air quality and related health issues. These areas are largely covered with asphalt pavements. Pollutant emissions from asphalts are generally investigated at 120–160 °C, temperatures that are reached during deposition of asphalt mixtures, but data at in-use temperatures are lacking. In the present work, we characterize and quantify in the laboratory emissions of nitrogen oxides (NOx = NO+NO2) by fresh and old asphalt mixtures under simulated atmospheric conditions. The impact of asphalt aging on NOx emissions is assessed, and the regeneration of NOx emissions by exposure of asphalt to the gas exhaust of a car is evidenced experimentally. With a numerical model of NOx emissions in Athens, Greece, we show that this previously unknown source accounts for up to 20% of NOx emissions in specific locations. Asphalt-covered surfaces can therefore significantly contribute to NOx urban pollution, and need to be included to emission inventories, and taken into account by models for a better understanding and description of factors influencing urban air quality.

An equilibrium model of NOx emission in gas turbine combustors

In this study, an equilibrium model for NOx emission in gas turbine combustors was developed. The model is based on the governing thermodynamic, materials, and energy balance equations. Results obtained from the model simulation were compared with operational data from a local gas turbine power plant. There was about a 3 to 6 percent deviation for the periods considered. Consequently, the model is reliable in the design of new combustors and in assessing the pollutant emission during operations. It was established that factors that affect NOx formation include flame temperature and quantity of excess air used. Furthermore, the results revealed that the most important parameter that governs the formation of NOx is flame temperature, as the formation of NOx rises exponentially with an increase in the combustor flame temperature.

Analysis of the Impact of Variable Speed Limits on Environmental Sustainability and Traffic Performance in Urban Networks

This work focuses on evaluating the potential of variable speed limits (VSLs) in a synthetic urban network to improve both environmental sustainability and traffic performance. The traffic system is modeled using the microscopic traffic simulator SUMO, and a physical fuel consumption and NOx emission model is used to assess the vehicles’ energy efficiency. Speed limits are controlled through a nonlinear model predictive control (NMPC) approach, in which the traffic evolution and fuel consumption are respectively predicted with a macroscopic traffic model, namely the cell transmission model (CTM), and a pre-calibrated artificial neural network (ANN). The results reveal that in transient phases between different levels of congestion, the proposed eco-VSL controller is faster to decongest the network, resulting in an improvement of the environmental sustainability and the traffic performance both in the controlled network, and at its boundary roads.

Modelling of SO2 and NOx Emissions from Coal and Biomass Combustion in Air-Firing, Oxyfuel, iG-CLC, and CLOU Conditions by Fuzzy Logic Approach

Chemical looping combustion (CLC) is one of the most advanced technologies allowing for the reduction in CO2 emissions during the combustion of solid fuels. The modified method combines chemical looping with oxygen uncoupling (CLOU) and in situ gasification chemical looping combustion (iG-CLC). As a result, an innovative hybrid chemical looping combustion came into existence, making the above two technologies complementary. Since the complexity of the CLC is still not sufficiently recognized, the study of this process is of a practical significance. The paper describes the experiences in the modelling of complex geometry CLC equipment. The experimental facility consists of two reactors: an air reactor and a fuel reactor. The paper introduces the fuzzy logic (FL) method as an artificial intelligence (AI) approach for the prediction of SO2 and NOx (i.e., NO + NO2) emissions from coal and biomass combustion carried out in air-firing; oxyfuel; iG-CLC; and CLOU conditions. The developed model has been successfully validated on a 5 kWth research unit called the dual fluidized bed chemical looping combustion of solid fuels (DFB-CLC-SF).

A new online optimization method for boiler combustion system based on the data-driven technique and the case-based reasoning principle

To adapt to the time-variability of boiler combustion systems, a new online combustion optimization method for boiler is proposed in this paper. The massive historical combustion data are preprocessed, and then an improved constrained fuzzy weighted rule is employed to extract combustion rules from historical combustion data. After that, an improved particle swarm optimization-based least square support vector machine is adopted to construct the dynamic mathematic model for boiler efficiency and NOx emission, respectively, and an improved multi-objective particle swarm optimization algorithm based on the well-construction dynamic mathematical model is proposed and applied to excavate deeply the combustion rules of boiler, and the optimization case library is constructed by integrating all combustion rules. At last similarity measure-based case-based reasoning method is employed to rapidly identify the well-performance similar cases from the optimization case library, which is helpful to complete the online combustion optimization. The effectiveness of proposed online optimization method for boiler is proved by applying it to an actual combustion process. The results showed that proposed online optimization method could take less time to gain a set of excellent operating solution, the NOx emission reduced by 9.236% on average and the boiler efficiency increased by 0.046% on average. Therefore, the proposed online combustion optimization method for boiler has the ability to realize the online combustion optimization of boiler.

Steady-State NOx Emission Model for Gas-Fired Heating and Hot Water Combi-Boilers with Factor Analysis and Artificial Neural Network

Steady-State NOx Emission Model for Gas-Fired Heating and Hot Water Combi-Boilers with Factor Analysis and Artificial Neural Network

Computationally Inexpensive 1D-CNN for the Prediction of Noisy Data of NOx Emissions From 500 MW Coal-Fired Power Plant

Coal-fired power plants have been used to meet the energy requirements in countries where coal reserves are abundant and are the key source of NOx emissions. Owing to the serious environmental and health concerns associated with NOx emissions, much work has been carried out to reduce NOx emissions. Sophisticated artificial intelligence (AI) techniques have been employed during the past few decades, such as least-squares support vector machine (LSSVM), artificial neural networks (ANN), long short-term memory (LSTM), and gated recurrent unit (GRU), to develop the NOx prediction model. Several studies have investigated deep neural networks (DNN) models for accurate NOx emission prediction. However, there is a need to investigate a DNN-based NOx prediction model that is accurate and computationally inexpensive. Recently, a new AI technique, convolutional neural network (CNN), has been introduced and proven superior for image class prediction accuracy. According to the best of the author’s knowledge, not much work has been done on the utilization of CNN on NOx emissions from coal-fired power plants. Therefore, this study investigated the prediction performance and computational time of one-dimensional CNN (1D-CNN) on NOx emissions data from a 500 MW coal-fired power plant. The variations of hyperparameters of LSTM, GRU, and 1D-CNN were investigated, and the performance metrics such as RMSE and computational time were recorded to obtain optimal hyperparameters. The obtained optimal values of hyperparameters of LSTM, GRU, and 1D-CNN were then employed for models’ development, and consequently, the models were tested on test data. The 1D-CNN NOx emission model improved the training efficiency in terms of RMSE by 70.6% and 60.1% compared to LSTM and GRU, respectively. Furthermore, the testing efficiency for 1D-CNN improved by 10.2% and 15.7% compared to LSTM and GRU, respectively. Moreover, 1D-CNN (26 s) reduced the training time by 83.8% and 50% compared to LSTM (160 s) and GRU (52 s), respectively. Results reveal that 1D-CNN is more accurate, more stable, and computationally inexpensive compared to LSTM and GRU on NOx emission data from the 500 MW power plant.

Nonlinear modeling of industrial boiler NOx emissions

ABSTRACT Pollutant emissions into the atmosphere are recognized as a significant problem in fossil fuel combustion. The pollution emission measurement in industrial boilers is difficult and expensive but fundamental for monitoring and controlling. Frequently continuous emissions monitoring (CEM) system is out of service or useless due to obsolescence, high maintenance cost, and so on or simply is not installed. When a system for measuring pollutant emissions is not available, an alternative method must be employed to get the pollutant emission value. According to the black-box model approach, this article describes the nonlinear modeling of NOx emissions from a utility boiler. Bayesian-Gaussian (BG), multilayer perceptron (MLP), and Volterra polynomial basis functions (VPBF) neural networks are developed for model benchmarking. Experimental data from a utility boiler was acquired in order to model definition and evaluation. The models process three boiler variables oxygen excess, fuel mass flow and flue gas recirculation gates for NOx emission estimation. Models with BG show better performance than models with MLP and VPBF for NOx prediction. Implications: The technology to control NOx emissions generated by combustion operates under strict regulations. In order to reduce NOx emissions, theoretical models of NOx generation have been studied extensively, including nitrogen chemistry and the dynamic flow of gas particles which is very complex. The new technology trends would require the continuous measurement of high precision NOx emissions to achieve further reductions in NOx emissions. Currently, NOx emissions are measured by a Continuous Emission Monitoring system, which turns out to be extremely expensive and difficult to maintain, so alternative low-cost solutions are desirable. Our contribution shows how algorithms based on different artificial intelligence techniques are viable and quality alternatives for the measurement of continuous NOx emissions. The NOx emissions models based on IA algorithms are viable alternatives that have versatility and self-tuning capacity due to the fact that they are based on boiler operation parameters which have valuable information few explored nowadays.

Development of Prediction Model for CO2 and NOx Emissions for Diesel Engine Vehicles by Considering Real Road Driving Environment

Development of Prediction Model for CO2 and NOx Emissions for Diesel Engine Vehicles by Considering Real Road Driving Environment