A novel multivariable grey prediction model and its application in forecasting coal consumption.

Area of Science:

• Energy Economics
• Environmental Science
• Mathematical Modeling

Background:

• Coal remains a critical global energy source, necessitating accurate consumption predictions.
• Effective coal consumption forecasting supports informed energy strategies and environmental policy development.
• Population and economic growth are key drivers of coal consumption.

Purpose of the Study:

• To develop a novel, more accurate model for predicting coal consumption.
• To address limitations of classical grey models in real-world scenarios.
• To evaluate the proposed model's performance against established forecasting techniques.

Main Methods:

• Establishment of a differential equation.
• Proposal of a novel multivariable Verhulst grey model (MVGM(1,N)) utilizing grey information differences.
• Extension of the classical single-variable Verhulst model to a multivariate framework.

Main Results:

• The MVGM(1,N) model demonstrates enhanced applicability by reducing reliance on saturated S-shaped and single-peak data.
• Simulation experiments in high coal consumption areas show MVGM(1,N) outperforms NLARX, ARIMA, and five classical grey models.
• The model accurately predicted coal consumption in Inner Mongolia and Gansu Provinces, China.

Conclusions:

• The novel MVGM(1,N) provides a more precise and robust method for coal consumption forecasting.
• The model's multivariate capability and reduced data dependency make it suitable for complex real-world applications.
• MVGM(1,N) is recommended for effective coal consumption prediction in energy planning and policy.