A novel electricity load forecasting based on probabilistic least absolute shrinkage and selection operator-Quantile regression neural network

Liu, Shumei; Chen, Huiwei; Liu, Peixue; Qin, Fuzhen; Fars, Ashk

doi:10.1016/j.ijhydene.2023.04.091

Cited by 4 publications

(1 citation statement)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At present, there is a lack of comprehensive analysis on the driving factors of carbon emissions at different scales. Liu et al [15] employed Lasso regression to screen the key factors influencing electricity utilization and to decrease the dimensionality of the modeling data. Similarly, Liu et al [16] used Lasso regression to discern the variables impacting China's natural gas demand.…”

Section: Introductionmentioning

confidence: 99%

The Prediction of Medium- and Long-Term Trends in Urban Carbon Emissions Based on an ARIMA-BPNN Combination Model

Hou,

Chen

2024

Energies

View full text Add to dashboard Cite

Urban carbon emissions are an important area for addressing climate change, and it is necessary to establish scientific and effective carbon emission prediction models to formulate reasonable emission reduction policies and measures. In this paper, a novel model based on Lasso regression, an ARIMA model, and a BPNN is proposed. Lasso regression is used to screen the key factors affecting carbon emissions, and the ARIMA model is used to extract the linear components of the carbon emission sequences, while the BPNN is used to predict the residuals of the ARIMA model. The final result is the sum of that from the ARIMA model and the BPNN. The carbon peak, carbon neutralization time, and emissions were analyzed under different scenarios. Taking Suzhou City as an example, the results show that the electricity consumption of the whole population is one of the key drivers of carbon emissions; the carbon emission prediction accuracy and stability of the ARIMA-BPNN combined model are better than those of the single model, which improves the reliability as well as the accuracy of the model’s prediction. However, under the constraints of the current policies, the goal of achieving carbon peaking by 2030 in Suzhou City may not be realized as scheduled. This novel carbon emission prediction model built was validated to provide a scientific basis for low-carbon urban development. This study presents an important reference value for predicting carbon emissions and formulating emission reduction measures in other cities.

show abstract

Section: Introductionmentioning

confidence: 99%