Accès libre

A Multi-Objective Optimization Framework for Low-Carbon Index Construction and Application in Green Finance

  
17 mars 2025
À propos de cet article

Citez
Télécharger la couverture

Figure 1.

Data preprocessing workflow from raw input to structured dataset.
Data preprocessing workflow from raw input to structured dataset.

Figure 2.

Flowchart of the hybrid algorithm, combining genetic algorithms and gradient-based refinement.
Flowchart of the hybrid algorithm, combining genetic algorithms and gradient-based refinement.

Figure 3.

Pareto front depicting trade-offs between carbon emissions and financial returns.
Pareto front depicting trade-offs between carbon emissions and financial returns.

Figure 4.

Convergence comparison of optimization algorithms.
Convergence comparison of optimization algorithms.

Figure 5.

Comparison of low-carbon index values under different weight scenarios.
Comparison of low-carbon index values under different weight scenarios.

Computational performance comparison_

Algorithm Runtime Iterations Convergence Speed
Proposed Hybrid Method 45 120 60
Genetic Algorithm (GA) 90 150 90
Gradient-Based Method 70 100 80
Algorithm Runtime Iterations Convergence Speed

Selected Pareto-optimal solutions_

Solution ID Carbon Emissions Financial Return Low-Carbon Index
A 100 12.5 0.85
B 150 14.2 0.80
C 200 15.8 0.75
D 250 17.0 0.70
E 300 18.5 0.65

Computed low-carbon index values across sectors_

Sector Carbon Intensity Renewable Energy Economic Growth Low-Carbon Index
Energy 150 45 3.2 0.78
Technology 120 50 4.0 0.82
Manufacturing 300 20 2.1 0.45
Transportation 250 30 2.5 0.58
Agriculture 180 35 2.8 0.67