Accurate Modeling of Carbon Emissions Under Urban Energy Consumption
DOI:
https://doi.org/10.13052/dgaej2156-3306.405614Keywords:
Energy consumption, gradient boosting decision tree, optimization algorithms, regularization techniques, carbon emissionsAbstract
Carbon emissions refer to greenhouse gases, mainly carbon dioxide, released into the atmosphere through human activities or natural carbon cycles. Accurate estimation of carbon emissions helps promote the transformation of social and economic systems toward sustainable development. However, existing methods for estimating carbon emissions have some drawbacks such as low accuracy and large errors. Therefore, this study utilizes two optimization algorithms, namely random search and recursive feature elimination, the stochastic gradient descent optimizer, and regularization techniques, to optimize and improve the gradient boosting decision tree algorithm. On this basis, a carbon emission estimation model for urban energy consumption is constructed. Experimental results show that the model achieves a precision of 98.3%, a recall of 96.3%, an F1 score of 97.9%, an efficiency of 95.7%, a precision error of 0.36%, and a false positive rate of 3.7%. All the above experimental data are superior to the three comparison models. Moreover, the research model still demonstrates strong robustness, generalization and applicability when facing different conditions and scenarios, fully demonstrating the superiority and feasibility of the research model. This provides a new approach for carbon emission estimation under urban energy consumption and contributes to promoting green economic development.
Downloads
References
Phoochinda W, Khoasitthiwong B. Guidelines in Local Alternative Energy Management of Communities in Thailand’s Central and Eastern Regions. Distributed Generation and Alternative Energy Journal, 2015, 30(2): 43–56.
Uddin I, Usman M, Saqib N, Makhdum M S A. The impact of geopolitical risk, governance, technological innovations, energy use, and foreign direct investment on CO2 emissions in the BRICS region. Environmental Science and Pollution Research, 2023, 30(29): 73714–73729.
Pata U K, Dam M M, Kaya F. How effective are renewable energy, tourism, trade openness, and foreign direct investment on CO2 emissions? An EKC analysis for ASEAN countries. Environmental Science and Pollution Research, 2023, 30(6): 14821–14837.
Hebbi C, Mamatha H. Comprehensive Dataset Building and Recognition of Isolated Handwritten Kannada Characters Using Machine Learning Models. Artificial Intelligence and Applications, 2023, 1(3):179–190.
Douiba M, Benkirane S, Guezzaz A, Azrour M. An improved anomaly detection model for IoT security using decision tree and gradient boosting. The Journal of Supercomputing, 2023, 79(3): 3392–3411.
Feng Q, Kim H S. Comparative Analysis of Risk Factor Weighting GBDT Methods for Enhancing the Accuracy of Flood Risk Assessment. Journal of the Korean Society of Hazard Mitigation, 2025, 25(2): 9–21.
Yang L, Zhu Q, Ge X, Guo L. Transmit antenna selection for millimeter-wave MIMO system based on GBDT. Journal of Communications and Information Networks, 2023, 8(1): 71–79.
Wang L, Chi J, Ding Y, Yao H Y, Guo Q, Yang H. Q. Transformer fault diagnosis method based on SMOTE and NGO-GBDT. Scientific Reports, 2024, 14(1): 7179–7190.
Guo G, Liu Y, Cao Z, Zhang D, Zhao X. Interpretable GBDT model-based multi-objective optimization analysis for the lateral inlet/outlet design in pumped-storage power stations. Journal of Hydroinformatics, 2024, 26(5): 1189–1205.
Zhao B, Cao W, Zhang J, Gao Y, Li B, Chen F. Fusion of GBDT and neural network for click-through rate estimation. Journal of Intelligent & Fuzzy Systems, 2025, 48(6): 835–847.
Liu J, Zhao H, Wang S, Liu G, Zhao J, Dong Z. Y. Real-time emission and cost estimation based on unit-level dynamic carbon emission factor. Energy Conversion and Economics, 2023, 4(1): 47–60.
Byrne B, Liu J, Bowman K W, Pascolini-Campbell M, Chatterjee A, Pandey S, Sinha S. Carbon emissions from the 2023 Canadian wildfires. Nature, 2024, 633(8031): 835–839.
Li T, Yu L, Ma Y, Duan T, Huang W, Zhou Y, Jiang T. Carbon emissions of 5G mobile networks in China. Nature Sustainability, 2023, 6(12): 1620–1631.
Sousa V, Bogas J A, Real S, Meireles I. Industrial production of recycled cement: Energy consumption and carbon dioxide emission estimation. Environmental Science and Pollution Research, 2023, 30(4): 8778–8789.
Pan B, Adebayo T S, Ibrahim R L, Al-Faryan M A S. Does nuclear energy consumption mitigate carbon emissions in leading countries by nuclear power consumption? Evidence from quantile causality approach. Energy & Environment, 2023, 34(7): 2521–2543.
Li H, Hao T, Li Z, Zhao E, Wang C, Xu L. Research on a self-coordinated optimization method for distributed energy resources targeting risk mitigation. Distributed Generation and Alternative Energy Journal, 2024, 39(3): 659–690.
Hiremath R, Moger T. Improving the DC-link voltage of DFIG driven wind system using modified sliding mode control. Distributed Generation and Alternative Energy Journal, 2023, 38(3): 715–742.
Priyatno A M, Widiyaningtyas T. A systematic literature review: recursive feature elimination algorithms. JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer), 2024, 9(2): 196–207.
Reza A A R, Rohman M S. Prediction Stunting Analysis Using Random Forest Algorithm and Random Search Optimization. Journal of Informatics and Telecommunication Engineering, 2024, 7(2): 534–544.
Qin W, Luo X, Zhou M C. Adaptively-accelerated parallel stochastic gradient descent for high-dimensional and incomplete data representation learning. IEEE Transactions on Big Data, 2023, 10(1): 92–107.
Ben Hamida S, Mrabet H, Chaieb F, Jemai A. Assessment of data augmentation, dropout with L2 Regularization and differential privacy against membership inference attacks. Multimedia Tools and Applications, 2024, 83(15): 44455–44484.
