Modeling of energy consumption and related GHG (greenhouse gas) intensity and emissions in Europe using general regression neural networks

Antanasijević, Davor; Pocajt, Viktor; Ristić, Mirjana; Perić-Grujić, Aleksandra A.

doi:10.1016/j.energy.2015.03.060

Cited by 57 publications

(19 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This outcome is consistent with other relevant literature, for example, MLR, MNLR, and GRNN models were used to forecast GHG emissions at national level and the results showed that the GRNN model gave the most preferable results [25]. When Firat et al [42] predicted the scour depth around circular bridge piers based on the data from various studies, the GRNN model performed superior to BPNN and MLR.…”

Section: Comparison Of Results Obtained By Modelssupporting

confidence: 76%

See 1 more Smart Citation

Estimating Carbon Dioxide (CO2) Emissions from Reservoirs Using Artificial Neural Networks

Chen

Huang

2018

Water

View full text Add to dashboard Cite

Freshwater reservoirs are considered as the source of atmospheric greenhouse gas (GHG), but more than 96% of global reservoirs have never been monitored. Compared to the difficulty and high cost of field measurements, statistical models are a better choice to simulate the carbon emissions from reservoirs. In this study, two types of Artificial Neural Networks (ANNs), Back Propagation Neural Network (BPNN) and Generalized Regression Neural Network (GRNN), were used to predict carbon dioxide (CO 2 ) flux emissions from reservoirs based on the published data. Input variables, which were latitude, age, the potential net primary productivity, and mean depth, were selected by Spearman correlation analysis, and then the rationality of these inputs was proved by sensitivity analysis. Besides this, a Multiple Non-Linear Regression (MNLR) and a Multiple Linear Regression (MLR) were used for comparison with ANNs. The performance of models was assessed by statistical metrics both in training and testing phases. The results indicated that ANNs gave more accurate results than regression models and GRNN provided the best performance. With the help of this GRNN, the total CO 2 emitted by global reservoirs was estimated and possible CO 2 flux emissions from a planned reservoir was assessed, which illustrated the potential application of GRNN.

show abstract

Section: Comparison Of Results Obtained By Modelssupporting

confidence: 76%

“…Artificial neural networks (ANNs) were frequently being used for the simulation of both water quality [23,24] and GHG emissions caused by energy consumption [25,26], and showed great potential for prediction. However, few research has been conducted to simulate GHG emissions from reservoirs using ANNs.…”

Section: Introductionmentioning

confidence: 99%

Estimating Carbon Dioxide (CO2) Emissions from Reservoirs Using Artificial Neural Networks

Chen

Huang

2018

Water

View full text Add to dashboard Cite

show abstract

“…The same technique was adopted by Antanasijević, et al [23] to predict energy use as well as the intensity of carbon footprint related to energy consumption. However, the difference between the two studies is that Antanasijević, et al's [23] study compared GRNN to MLR (Multiple Linear Regression) to find the best prediction method using data from 26 European countries for the period of 2004 to 2012. The results exposed cost function (i.e.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Nuroğlu [14] argued that, based on the results, the prediction technique of the neural network is superior to the econometric method. There are so many instances in empirical studies that ANN has been applied to predict various kinds of variables with different explanatory variables based on country-level and cross-section of countries [15,[19][20][21][22][23][24]. A study in Turkey by Özceylan [25] predicted carbon dioxide emissions; while energy consumption was predicted by Azadeh, et al [26] in Iran; Geem and Roper [27] energy demand in South Korea; Ekonomou [28] energy consumption in Greece and Pao [29] total energy use in Taiwan.…”

Section: Introduction mentioning

confidence: 99%

Prediction of Potential Carbon Dioxide Emissions of Selected Emerging Economies Using Artificial Neural Network

Appiah¹,

Du²,

Appah³

et al. 2018

JESE-A

View full text Add to dashboard Cite

Tackling future global emissions of carbon dioxide is a daunting task. Different black box models have been used to determine the trajectories of CO 2 emissions and other carbon stocks. Trajectories are important because climate modelers use them to project future climate under higher atmospheric CO 2 concentrations. In this paper, fully connected two-layer feed-forward neural network with tangent activation function that comes with hidden neurons as well as linear output neurons was used. The study applied classical nonlinear least squares algorithm such as LM (Levenberg-Marquardt), to predict potential emissions of selected emerging economies. Building the model on the basis of input variables such as crop production, livestock production, trade imports, trade exports, economic growth, renewable and nonrenewable energy consumption. These variables are considered to affect the ecosystems of high rising economic power states. The main idea is to ensure that emerging economies have a clear understanding of expected future emissions so that appropriate measures can be implemented to mitigate its impact. Data for the analysis were obtained from 1971 to 2013 from World Development Indicators and FAOSTAT database. Results indicate an achievement of training performance at epoch 11 when the value of the MSE (Mean Square Error) is 0.0003345 which indicates that the model errors are less than 0.05. Hence, the study concluded that the applied model is capable of predicting potential carbon dioxide emissions in emerging economies with the greatest precision.

show abstract

“…In the above notation, the smoothing factor indicates the width of the Gaussian curve for the calculated joint pdf (Antanasijevic et al 2015). Indeed, the optimization of the smoothing factor is believed to be among the prominent computing tasks of the GRNN development (Huang and Williamson 1994).…”

Section: Fig 12mentioning

confidence: 99%

A pragmatic approach for optimizing gas lift operations

Garrouch

Al-Dousari

Al-Sarraf

2019

J Petrol Explor Prod Technol

View full text Add to dashboard Cite

The oil flow rate in a single vertical well undergoing gas lift operations is complicated by three factors: (1) The flow is driven by gas injection, in addition to the fluid flow potential gradient applied along the well, (2) the well is interfaced with a porous and permeable reservoir contributing with a fluid feed, and (3) the wellbore geometry may consist of concentric pipes of varying diameters and lengths, rather than a single-diameter pipe. Dimensional analysis is applied to this complex, highly nonlinear production problem, in order to develop empirical models for predicting the optimal gas injection rate and the maximum oil production rate that may be produced from continuous gas lift operations. Two pairs of coupled dimensionless groups are revealed. The first pair consists of a dimensionless pressure drop (π 1) adjusted to the complex wellbore geometry, and a dimensionless ratio of kinetic to viscous forces (π 2) which accounts for the porous medium feed. A constructed database for 388 vertical wells producing by continuous gas lift operations has been used to validate the dimensionless groups. A power-law relation is revealed between the dimensionless groups π 1 and π 2 , allowing to construct an analytical model for predicting the maximum oil production rate that corresponds to the optimal gas injection rate. The second pair consists of two groups denoted χ 1 and χ 2. The group χ 1 is a dimensionless pressure drop with adjustment being augmented to account for the temperature effects on gas flow. Similar to π 2 , the dimensionless group χ 2 is a ratio of kinetic to viscous forces, adjusted to account for the porous medium feed. However, χ 2 is a function of the injection rate, instead of the oil production rate. Likewise, a power-law relation is revealed between χ 1 and χ 2 , allowing to construct an analytical model for predicting the optimal gas injection rate. All power-law relations yield high correlation coefficients when the validation data are segregated according to a discrete productivity index. The analytical models developed by applying dimensional analysis appear to capture the physical controls of gas lift operations. Intuitively, the optimal gas injection rate depends on the pressure gradient along the pipe, the wellbore geometry, the temperature conditions at the bottom of the well and in the stock-tank, the oil density, and on the productivity index. Similarly, the maximum oil production rate, corresponding to the optimal gas injection rate, depends on the pressure gradient along the pipe, the wellbore geometry, the oil density, the productivity index which is implicitly affected by the oil permeability, and viscosity. Unlike multivariate nonlinear regression analysis, the application of dimensional analysis for deriving the analytical models, presented in this study, does not require a presumed functional relationship. In retrospect, dimensional analysis evades the guessing process associated with nonlinear regression analysis.

show abstract

Modeling of energy consumption and related GHG (greenhouse gas) intensity and emissions in Europe using general regression neural networks

Cited by 57 publications

References 32 publications

Estimating Carbon Dioxide (CO2) Emissions from Reservoirs Using Artificial Neural Networks

Estimating Carbon Dioxide (CO2) Emissions from Reservoirs Using Artificial Neural Networks

Prediction of Potential Carbon Dioxide Emissions of Selected Emerging Economies Using Artificial Neural Network

A pragmatic approach for optimizing gas lift operations

Contact Info

Product

Resources

About