Emission Prediction Model Research Articles

A hybrid deep neural network model for NOx emission prediction of heavy oil-fired boiler flames

Accurate NOx emission monitoring is essential for an in-depth understanding of the combustion state. However, establishing an accurate emission prediction model based on the traditional data-driven method is difficult, limited by low robustness and insufficient labeled data. To mitigate these limitations, this study proposes a hybrid deep neural network model for NOx emission prediction. In this hybrid model, an adversarial denoising autoencoder (ADAE) is established to extract flame deep features, and then the least support vector regression (LSSVR) is utilized to analyze the extracted features for predicting NOx emission. A novel training strategy that includes adversarial mechanisms, denoising coding and dropout is introduced to enhance the feature learning ability. The feasibility of the hybrid model ADAE-LSSVR is verified by the 4.2 MW heavy oil-fired boiler flame images, and the model hyper-parameters are optimized for higher prediction performance. Experimental results demonstrated that the ADAE can automatically extract robust features from raw flame images without manual intervention. Moreover, the proposed ADAE-LSSVR model provides satisfactory prediction accuracy with a correlation coefficient (R2) of 0.97, outperforming other state-of-art models.

Studies on the Influencing Factors of Gas Emission in a Steeply Inclined and Ultrathick Coal Seam Working Face.

To determine the factors affecting the gas emission in the working face during the horizontal sublevel mining of steeply inclined and ultrathick coal seams (SIUTCSs), the gas emission sources were identified and evaluated using the analytic hierarchy process (AHP), and the gas emission quantity was calculated using a prediction model. Sobol sensitivity analysis was then conducted on the influencing factors involved in the model to determine their influences on the working face gas emission. The AHP analysis found that the pressure relief area in the lower section is the main gas emission source of the working face, and the adjacent coal seams, old goaf, and rear goaf are not. The gas emission prediction model exhibits good accuracy. The calculation results suggest that the gas released from the lower section coal body accounts for 44.16-50.44% of the total gas emission quantity of the working face, and thus, it is the main gas emission source. The Sobol sensitivity analysis reveals that the dip angle of the coal seam has the greatest influence on the absolute gas emission quantity of the working face with a significantly larger sensitivity than those of other factors. The comprehensive sensitivity data analysis also suggests that the lower section coal body is the major contributor to the gas emission of the working face. Our work further puts forward a technical system of ultrahigh pressure hydraulic "drilling-slitting-pressing-draining" integrated antireflection enlarged gas extraction for controlling the gas emission from the coal body at the bottom of the SIUTCS. The engineering test demonstrates that this system can increase the permeability of the coal body and significantly improve the gas extraction efficiency of the coal body in the study area.

Prediction of Shanghai Electric Power Carbon Emissions Based on Improved STIRPAT Model

Energy is the bridge connecting the economy and the environment and electric energy is an important guarantee for social production. In order to respond to the national dual-carbon goals, a new power system is being constructed. Effective carbon emission forecasts of power energy are essential to achieve a significant guarantee for low carbon and clean production of electric power energy. We analyzed the influencing factors of carbon emissions, such as population, per capita gross domestic product (GDP), urbanization rate, industrial structure, energy consumption, energy structure, regional electrification rate, and degree of opening to the outside world. The original Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) model was improved, and the above influencing factors were incorporated into the model for modeling analysis. The ridge regression algorithm was adopted to analyze the biased estimation of historical data. The carbon emission prediction model of Shanghai electric power and energy based on elastic relationship was established. According to the “14th Five-Year” development plan for the Shanghai area, we set up the impact factor forecast under different scenarios to substitute into the forecast models. The new model can effectively assess the carbon emissions of the power sector in Shanghai in the future.

Investigation of sniffer technique on remote measurement of ship emissions: A case study in Shanghai, China

Shipping emissions have aroused wide concern in the world. In order to promote the implementation of emission regulations, this study develop a ship based sniffing technique to perform remote measurement of the SO2, NOx and CO2 from ships entering and leaving Shanghai port at the open sea. The ship emission prediction model, Smoke diffusion model and source identification model were developed to automatically analyze the emission data and screen the object ship source based on Automatic Identification System (AIS) system. The fueling documents of the detected ship were obtained from maritime sector and the results precision of the sniffer technique was evaluated by comparing the measured Fuel sulfur content (FSC) with actual value deduced from fueling documents. The influences of wind speed and direction, object ship parameters and monitoring distance on the identification of object ship and accuracy of the calculated FSC were thoroughly investigated and the corresponding correction factors under different conditions were deduced. The modified emission factor ratio of CO2 to NOx were proposed in order to improve the accuracy. It is demonstrated that with wind speed higher than 2 m/s and test distance shorter than 400m, the sniffer technique exhibit high efficiency and accuracy for the remote emissions measurement of ship upwind with detection rate higher than 90% and test error of FSC below 15%. To reduce the influence of the wind direction, at least two sniffer systems were required to guarantee that at least one station is in the downwind of the ship lane. Based on the results and discussion, a novel sniffer monitoring system with two buoy based sniffing stations placed close to each side of the ship lane far off shore was proposed to realize the remote monitoring of ship emissions.

Prediction of Gas Emissions in the Working Face Based on the Desorption Effects of Granular Coal: A Case Study

The aim of the study in this paper is to establish a prediction model of gas emission in the working face. The gas desorption variation characteristics of coal with different particle sizes were assessed using physical tests and based on the coal body of No. 2 coal seam in Wangjialing Coal Mine, Shanxi, China, to reveal the influence law of coal particle size on coal gas desorption. The gas desorption characteristics in the working face, the law of gas emission of coal cutting, coal caving, coal wall, and remnant coal in the goaf of the production process were then analyzed after establishing a gas emission prediction model based on the particle size of the coal. The accuracy of the gas emission prediction model was finally validated through actual measurement of the coal particle size distribution and gas emission in the test working face. The results of the current study show that the coal particle size is negatively correlated with the gas desorption capacity within a certain range. The initial desorption intensity of the coal gas decreased with an increase in the coal particle size. However, the initial gas desorption intensity and attenuation coefficient of gas emission were constant after a certain level of increase in the coal particle size. It was found that the average error between the gas emission prediction model and the actual gas emission data in the mining process was 5.29% based on the desorption characteristics of granular coal. Therefore, the established gas emission prediction model can characterize the law of gas emission in the actual production process more effectively. Furthermore, it provides reliable support for the prediction and control of gas emissions from the goaf under the condition of fully mechanized mining with top coal caving.

Effects of ruminal protozoa on methane emissions in ruminants—A meta-analysis

The effects of different ruminal protozoa (RP) on CH4 emissions from ruminants were evaluated in a meta-analysis, using 64 publications reporting data from 79 in vivo experiments. Experiments included in the database reported CH4 emissions (g/d) and total RP (TRP, log10 cells/mL) from the same group of animals. The relationship between CH4 emissions and RP (TRP, entodiniomorphids, and isotrichids), and TRP-, entodiniomorphid-, and isotrichid-based CH4 emission prediction models, were evaluated as mixed models with experiment as a random effect and weighted by the reciprocal of the standard error of the mean and centered around one. Positive associations existed between TRP and isotrichids with CH4 emissions but not between entodiniomorphids and CH4 emissions. A reduction in CH4 emissions was observed, averaging 7.96 and 4.25 g/d, per log unit reduction in TRP and isotrichid concentrations, respectively. Total RP and isotrichids were important variables in predicting CH4 emissions from ruminants. Isotrichid CH4 prediction model was more robust than the TRP, evidenciated by lower predicted sigma hat study (%), and error (%), and with higher concordance correlation coefficient. Both TRP and isotrichid models can accurately predict CH4 emissions across different ruminant types, as shown by the low square root of the mean square prediction error, with 6.59 and 4.08% of the mean of root of the mean square prediction error in the TRP and isotrichid models, respectively. Our results confirm that isotrichids are more important than entodiniomorphids in methanogenesis. Distinguishing these 2 populations yielded a more robust CH4 prediction model than combining them as total protozoa.

Quantitative analysis of impact factors and scenario prediction of energy related carbon emissions at county level

ABSTRACT The key to coping with climate change is to control carbon emissions from energy consumption. Scientific prediction of energy consumption carbon emissions based on influencing factors is of great significance to the determination of carbon control aim and emission reduction strategies. Given the lack of previous studies on county-level carbon emissions, this paper proposed a systematic approach to study the influencing factors of county-level energy consumption carbon emissions and to predict future emissions. Firstly, the annual energy consumption carbon emissions were calculated based on the method proposed by the Intergovernmental Panel on Climate Change (IPCC). Then the expanded Kaya equation and existing research were combined to select influencing factors for the establishment of the optimal Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) model, which was used to quantitatively analyze the influencing factors of carbon emissions from energy consumption at the county level. Finally, the emission reduction aims and low-carbon strategies were determined based on scenario analysis. The method was applied to Changxing, a typical county with large energy consumption and carbon emissions. Based on 16 years of data, the STIRPAT carbon emission prediction model was established and the forecast results of future emissions under three different scenarios were obtained. The results indicated that population size, industrial structure, and affluence degree were the three most influential factors, and the influence degree of each factor was quantified to support targeted low-carbon strategies for county-level cities.

A daily carbon emission prediction model combining two-stage feature selection and optimized extreme learning machine.

Global warming caused by increased carbon emissions is a common challenge for all mankind. Facing the unprecedented pressure of carbon emission reduction, it is particularly important to grasp the dynamics of carbon emission in time and accurately. This paper proposes a novel daily carbon emission forecasting model. Firstly, the daily carbon emission data is decomposed into a series of completely noise-free mode functions by improved complete ensemble empirical mode decomposition method with adaptive noise (ICEEMDAN). Then, a two-stage feature selection method composed of partial autocorrelation function (PACF) and ReliefF is applied to select appropriate input variables for the next prediction process. Finally, the extreme learning machine optimized by improved sparrow search algorithm (ISSA-ELM) is used to predict. The empirical results show that the proposed two-stage feature selection method can further improve the prediction accuracy. After two-stage feature selection, the values of R2, MAPE, and RMSE were improved by 0.55%, 30.23%, and 28.46%, respectively. It can also be found that ISSA has good optimization performance. By combining with ISSA, R2, MAPE, and RMSE improved by 7.60%, 31.97%, and 44.79%, respectively. Therefore, the proposed model can provide a valuable reference for the formulation of carbon emission reduction policies and future carbon emission prediction research.

Carbon emission and thermal comfort prediction model for an office building considering the contribution rate of design parameters

In recent years, increasing climate change has triggered the development of various techniques and strategies for predicting building energy consumption and thermal comfort. However, many features could affect prediction efficiency, and the coupling of appropriate learning algorithms with different sampling methods and input parameters requires further research. In this study, we evaluated 25 features of an office building model. Then, the comprehensive effects of five different sampling methods and cumulative contribution rates (CCR) of input parameters on the prediction performance of building carbon emission (BCE) and indoor discomfort hour (IDH) were explored in 10 machine learning algorithms. The results indicated that the Sobol sampling method could achieve the best prediction effect in the combinations of different contribution rates of features and machine learning algorithms. Meanwhile, an artificial neural network (ANN) was the best learning algorithm when the CCR was 100%. However, the optimal machine learning method corresponding to each CCR stage differed. When the CCR was reduced to 50%, only three influence factors were considered most important, and K-nearest neighbors (KNN) was the best prediction algorithm in this scenario, indicating that the input parameters were reduced by 88%, and the R2 value was only reduced by 6.1%. Therefore, we proposed a new strategy idea for the research of building performance prediction by determining the most important building impact factors and suitable machine learning algorithms, which could simplify the prediction process and improve prediction efficiency while reducing building parameters.

Data-Based Engine Torque and NOx Raw Emission Prediction

Low accuracy is the main challenge that plagues the application of engine modeling technology at present. In this paper, correlation analysis technology is used to analyze the main influencing factors of engine torque and NOx (nitrogen oxides) raw emission performance from a statistical point of view, and on this basis, the regression algorithm is used to construct the engine torque and NOx emission prediction model. The prediction RMSE between engine torque prediction value and true value reaches 4.6186, and the torque prediction R2 reaches 1.00. Prediction RMSE between NOx emission prediction value and true value reaches 67.599, and NOx emission prediction R2 reaches 0.99. When using the new WHTC data for model prediction verification, the RMSE between the engine torque predicted value and true value reaches 4.9208, and the prediction accuracy reaches 99.60%, the RMSE between NOx emission prediction value and true value reaches 72.38, and the prediction accuracy reaches 99.2%, indicating that the model is relatively accurate. The evaluation result of the ambient temperature impact on torque shows that ambient temperature is positively correlated with engine torque.

Carbon Emission Prediction Model and Analysis in the Yellow River Basin Based on a Machine Learning Method

Excessive carbon emissions seriously threaten the sustainable development of society and the environment and have attracted the attention of the international community. The Yellow River Basin is an important ecological barrier and economic development zone in China. Studying the influencing factors of carbon emissions in the Yellow River Basin is of great significance to help China achieve carbon peaking. In this study, quadratic assignment procedure regression analysis was used to analyze the factors influencing carbon emissions in the Yellow River Basin from the perspective of regional differences. Accurate carbon emission prediction models can guide the formulation of emission reduction policies. We propose a machine learning prediction model, namely, the long short-term memory network optimized by the sparrow search algorithm, and apply it to carbon emission prediction in the Yellow River Basin. The results show an increasing trend in carbon emissions in the Yellow River Basin, with significant inter-provincial differences. The carbon emission intensity of the Yellow River Basin decreased from 5.187 t/10,000 RMB in 2000 to 1.672 t/10,000 RMB in 2019, showing a gradually decreasing trend. The carbon emissions of Qinghai are less than one-tenth of those in Shandong, the highest carbon emitter. The main factor contributing to carbon emissions in the Yellow River Basin from 2000 to 2010 was GDP per capita; after 2010, the main factor was population. Compared to the single long short-term memory network, the mean absolute percentage error of the proposed model is reduced by 44.38%.

Construction of Game Model between Carbon Emission Minimization and Energy and Resource Economy Maximization Based on Deep Neural Network.

Under this background, this paper tries to find countermeasures and ways for carbon reduction by observing and analyzing the influencing factors of carbon emissions, designing ways to minimize carbon emissions and maximize resources and energy. In view of the above problems, the carbon emission prediction research is closely combined with the research of deep neural network, the carbon emission prediction models based on deep neural network are established, respectively, and the game theory is introduced to maximize the resource economy. Based on the analysis of the cost of energy resources, this paper puts forward a model based on game theory and makes an overall planning of the bidding online auxiliary decision-making system in combination with the actual market demand. Build a big data analysis platform based on the Internet of things, collect the data related to carbon emission for normalization, analyze the influencing factors related to carbon emission by using the principal component analysis method, select the data with higher connection value, and take the time series data as the input of the deep neural network for simulation verification. The simulation results show that the game model of carbon emission minimization and energy resource economic maximization based on deep neural network can effectively improve the economic maximization of energy resources and reduce carbon emissions.

A novel short-term carbon emission prediction model based on secondary decomposition method and long short-term memory network.

Grasping the dynamics of carbon emission in time plays a key role in formulating carbon emission reduction policies. In order to provide more accurate carbon emission prediction results for planners, a novel short-term carbon emission prediction model is proposed. In this paper, the secondary decomposition technology combining ensemble empirical mode decomposition (EEMD) and variational mode decomposition (VMD) is used to process the original data, and the partial autocorrelation function (PACF) is applied to select the optimal model input. Then, the long short-term memory network (LSTM) is chosen for prediction. The secondary decomposition algorithm is innovatively introduced into the field of carbon emission prediction, and the empirical results illustrate that the secondary decomposition technology can further improve the prediction accuracy. Combined with the secondary decomposition, the R2, MAPE, and RMSE of the model are improved by 2.20%, 43.08%, and 36.92% on average. And the proposed model shows excellent prediction accuracy (R2 = 0.9983, MAPE = 0.0031, RMSE = 118.1610) compared with other 12 comparison models. Therefore, this model not only has potential value in the formulation of carbon emission reduction plans, but also provides a valuable reference for future carbon emission forecasting research.

Study on control-oriented emission predictions of PPCI diesel engine with two-stage fuel injection

Model predictive control (MPC) is an effective technology used to improve the control accuracy of low temperature combustion engines. The main purpose of this study is to establish a control-oriented emission prediction model based on fuel injection parameters, which provides the basis for the development of MPC controller for PPCI engines. In this study, the combustion model and backpropagation (BP) neural network emission prediction model of a partially premixed compression ignition (PPCI) diesel engine were established, through which the real-time combustion and emissions can be realized by inputting two-stage fuel injection parameters. The Wiebe function linearization method was proposed for the combustion model to achieve high-precision reconstruction of combustion at different fuel injection parameters. The nitrogen oxide (NOx) and particulate matter (PM) emissions were predicted using the combustion characteristics calculated by the combustion model. The results show that the established combustion model can accurately reconstruct the heat release rate and cylinder pressure of PPCI combustion, and the maximum error between the calculated and test values of the combustion parameters is less than 9%. The three-layer BP neural network can achieve high-precision predictions of NOx and PM emissions, with a prediction accuracy greater than 95%. Under the 300th continuous transient seconds, the cumulative prediction error of NOx was 2%, and the cumulative prediction error of PM was 4.9%, indicating high accuracy. This provides the basis for developing MPC controllers for PPCI engines.

Dioxin emission prediction based on improved deep forest regression for municipal solid waste incineration process

Dioxin (DXN) emission concentration is an important environmental indicator in the municipal solid waste incineration (MSWI) process. The prediction model of DXN emission can be used for pollution control to realize actual requirements of operation optimization. Therefore, a DXN emission concentration prediction model based on improved deep forest regression (ImDFR) is proposed in this study. A feature reduction layer based on out-of-bagging error is first introduced into the ImDFR to eliminate redundant variables and feed all confidence information on DXN emission into the feature enhancement layer of the MSWI process. A deep ensemble stacking model is subsequently built to depict deep features and increase diversity and accuracy using random forests, completely random forests, GBDT, and XGBoost as subforests. Finally, the predicted value of the DXN prediction model is determined in the decision layer. The DXN emission prediction model is verified using actual historical data of two incinerators operated with a daily processing capacity of 800 tons. The experimental results showed that the proposed prediction model presents higher accuracy and better generalization ability than state-of-the-art models.

A Review on the CO2 Emission Reduction Scheme and Countermeasures in China’s Energy and Power Industry under the Background of Carbon Peak

To reach the peak of carbon emission in China, the energy and power industry has the most arduous task and the heaviest responsibility. It should not only ensure efficient economic development, but also complete the arduous task of energy conservation and emission reduction. It is the main force in helping reach the peak of carbon emission. Taking the achievement of carbon peak in China’s power industry as the research object, this paper utilizes time series analyses to establish CO2 emission prediction models for China and its power industry under two scenarios: with and without a carbon peak target. The paper analyzes the current status of achieving carbon peak in China’s power industry and puts a forward CO2 emission reduction scheme for China and its power industry in the future. On this basis, countermeasures for China’s power industry to deal with carbon peak are explored.

Prediction of Carbon Emission Peak and Selection of Development Path in Gansu Province Based on the Constructed STIRPAT Model

The ridge regression method was applied to analyze the factors affecting carbon emission in Gansu from 2000 to 2019 based on the estimated carbon emission data, constructed carbon emission prediction model and predicted the trend of carbon emissions of Gansu Province under six scenarios. The results indicated that the overall trend of carbon emissions in Gansu Province from 2000 to 2019 is growing; and the carbon emission in Gansu was significantly affected by the size of population, industrial structure, and urbanization rate; Gansu Province could achieve carbon peak in 2040 under six development models; The peak of carbon emission in Gansu Province is proportional to the economic growth rate and inversely proportional to the emission reduction efforts, the faster its economic development rate, the more sensitive the peak size is to the emission reduction effort, and the smaller the emission reduction effort, the more sensitive the peak size is to the economic development rate; Low peaks are generally peaked early, and high peaks are generally peaked late. Based on the above analysis, it is believed that the more suitable carbon peak development path for Gansu Province is the he medium economic growth with medium emission reduction model.

Influencing Factors and Forecasting of Carbon Emissions in China's Logistics Industry——Based on GM-ARIMA Model

As an important industry in China's national economy, the logistics industry has an increasing proportion of carbon emissions year by year. In order to achieve China's carbon emission reduction target, it is urgent to study the emission reduction path of the logistics industry. In this paper, GM and ARIMA are combined in a weighted way to improve the prediction accuracy, uses the GM-ARIMA method to select 12 influencing factors such as population, per capita GDP, and per capita urban road area to establish a carbon emission prediction model for China's logistics industry. And take the carbon emission data from 2000 to 2020 as an example to conduct empirical research. The results show that: Compared with the GM (1,1), the GM-ARIMA prediction has a 4.33% reduction in error, which has a better prediction effect on the carbon emissions of the logistics industry. The carbon emission of China's logistics industry will peak at 967.45Mt CO2 in 2028. To achieve the carbon peaking target, China should introduce more emission reduction policies related to logistics.

Novel Economy and Carbon Emissions Prediction Model of Different Countries or Regions in the World for Energy Optimization Using Improved Residual Neural Network

Novel Economy and Carbon Emissions Prediction Model of Different Countries or Regions in the World for Energy Optimization Using Improved Residual Neural Network

A Combined Nox Emission Prediction Model Based on Semi-Empirical Model and Black Box Model

A Combined Nox Emission Prediction Model Based on Semi-Empirical Model and Black Box Model