A Novel Approach for Predicting CO2 Emissions in the Building Industry Using a Hybrid Multi-Strategy Improved Particle Swarm Optimization–Long Short-Term Memory Model

Energies, Journal Year: 2024, Volume and Issue: 17(17), P. 4379 - 4379

Published: Sept. 1, 2024

The accurate prediction of carbon dioxide (CO2) emissions in the building industry can provide data support and theoretical insights for sustainable development. This study proposes a hybrid model predicting CO2 that combines multi-strategy improved particle swarm optimization (MSPSO) algorithm with long short-term memory (LSTM) model. Firstly, (PSO) is enhanced by combining tent chaotic mapping, mutation least-fit particles, random perturbation strategy. Subsequently, performance MSPSO evaluated using set 23 internationally recognized test functions. Finally, predictive MSPSO-LSTM assessed from Yangtze River Delta region as case study. results indicate coefficient determination (R2) reaches 0.9677, which more than 10% higher BP, LSTM, CNN non-hybrid models demonstrates significant advantages over PSO-LSTM, GWO-LSTM, WOA-LSTM models. Additionally, mean square error (MSE) 2445.6866 Mt, absolute (MAE) 4.1010 both significantly lower those Overall, high accuracy industry, offering robust development industry.

Language: Английский

---

Carbon peak prediction for differentiated cities from a low-carbon perspective: Key factors, scenario analysis, and low-carbon pathways

Ecological Indicators, Journal Year: 2024, Volume and Issue: 167, P. 112629 - 112629

Published: Sept. 21, 2024

Language: Английский

---

Applications of Fractional Order Logistic Grey Models for Carbon Emission Forecasting

Fractal and Fractional, Journal Year: 2024, Volume and Issue: 8(3), P. 145 - 145

Published: Feb. 29, 2024

In recent years, global attention to carbon emissions has increased, becoming one of the main drivers climate change. Accurate prediction emission trends in small and medium-sized countries scientific regulation can provide theoretical support policy references for effective rational use energy promotion coordinated development energy, environment, economy. This paper establishes a grey model using classical Logistic mathematical determined environment investigate system. At same time, we basic principle fractional-order accumulation establish with obtain parameter estimation time-response equation new by solving through theory related operators. The particle swarm optimization algorithm is used complete process order fractional optimal order. Then, applied predict five medium-emission countries: Ethiopia, Djibouti, Ghana, Belgium, Austria. shows better advantages validity analysis process, simulation results indicate that proposed this stronger stability accuracy than other comparative models, proving model’s validity. Finally, forecast these years 2021–2025, are analyzed, relevant recommendations made.

Language: Английский

---

Application of improved graph convolutional networks in daily-ahead carbon emission prediction

Frontiers in Energy Research, Journal Year: 2024, Volume and Issue: 12

Published: April 5, 2024

With the increasing complexity of power systems and proliferation renewable energy sources, task calculating carbon emissions has become increasingly challenging. To address these challenges, we developed a new method for predicting emission factors. Bayesian optimization technique graphical convolutional networks with long- short-term network (BO-TGNN) is used to predict system. The aims quickly day-ahead system nodes enhanced feature extraction optimized training hyperparameters. effectiveness proposed demonstrated through simulation tests on three different using four deep learning algorithms. provides tailored solution evolving needs reduction efforts significant step forward in addressing calculations modern systems.

Language: Английский

---

Data-Driven Approaches for Achieving Carbon Neutrality: Predictive Models for Reducing CO2 Emissions and Enhancing Industrial Sustainability

Published: Jan. 1, 2024

In response to the escalating challenges posed by climate change and industrial inefficiency, this thesis presents a comprehensive investigation aimed at advancing predictive modeling of global CO2 emissions enhancing operational efficiency in steel manufacturing through Electric Arc Furnace (EAF) temperature optimization. Leveraging rich dataset sourced from World Development Indicators database alongside meticulously curated specific EAF operations, our study applies an innovative blend econometric machine learning techniques, including Pooled Ordinary Least Squares (Pooled OLS), Random Effects (RE), Fixed (FE), Seasonal Autoregressive Integrated Moving Average with Exogenous Variables (SARIMAX) models. The objective is twofold: refine emission forecasts establish reliable model for predicting flat bath production, critical determinant energy product quality. Our analysis elucidates complex dynamics governing emissions, identifying key factors such as renewable consumption, GDP per unit use, total greenhouse gas significant determinants. These insights not only contribute academic discourse on environmental sustainability but also provide solid foundation policymakers devise more effective strategies reduction. Concurrently, realm manufacturing, breaks new ground harnessing data predict unprecedented accuracy. This advancement holds implications conservation optimization, addressing urgent need practices. bridges gap between theoretical research practical application sets benchmark utilization data-driven approaches science engineering. By offering detailed comparison techniques their prowess, it guides future directions underscores potential sophisticated analytical methods tackling some most pressing challenges. Ultimately, role achieving sustainable future, providing valuable that can inform both policy process

Language: Английский

---