A grey NGM(1,1, k) self-memory coupling prediction model for energy consumption prediction

Area of Science:

• Energy Systems Analysis
• Predictive Modeling
• Grey System Theory

Background:

• Accurate energy consumption prediction is crucial for governments and investors.
• Existing prediction techniques face challenges with non-homogeneous exponential data sequences common in energy systems.
• The sensitivity of traditional grey models to initial values limits their predictive performance.

Purpose of the Study:

• To introduce a novel grey NGM(1,1,k) self-memory coupling prediction model.
• To enhance predictive accuracy for energy consumption data.
• To overcome the limitations of traditional grey prediction models.

Main Methods:

• Development of a grey NGM(1,1,k) self-memory coupling model.
• Integration of the self-memory principle with the grey NGM(1,1,k) model.
• Application of the model to China's total energy, coal, and electricity consumption data.

Main Results:

• The proposed NGM(1,1,k) self-memory coupling model demonstrated superior prediction performance.
• The model effectively utilizes multi-time historical data and captures stochastic fluctuation tendencies.
• Results surpassed those reported in existing literature for the same datasets.

Conclusions:

• The NGM(1,1,k) self-memory coupling model offers enhanced predictive capabilities for energy consumption.
• This approach effectively addresses the initial value sensitivity issue of traditional grey models.
• The study contributes to grey prediction theory and expands its practical applications in energy forecasting.