Forecasting nuclear energy consumption in China and America: An optimized structure-adaptative grey model

Forecasting nuclear energy accurately is vital to ensure reliable electricity supply and alleviate environmental degradation problems. However, it is difficult to model the nuclear time series due to its complexity, nonlinearity, and uncertainty. To this end, this paper put forward an optimized structure adaptive grey model by theoretically providing the generalized time response function and accurately modifying the background value based on Simpson's rule. Moreover, Monte-Carlo Simulation and probability density analysis (PDA) are employed to analyze the proposed model's merits, revealing its robustness and reliability. For validation and verification purposes, the optimized model is implemented to predict nuclear energy consumption in China and America, compared to seven benchmark models involving other prevalent grey models, conventional econometric technology, and artificial intelligence. Experimental results from two cases consistently demonstrate that the novel technique significantly outperforms the competitors from two different perspectives of PDA and level accuracy. Besides, further discussion over different forecasting horizons reveals that this new model can still deliver accurate forecasts with solid robustness and high reliability. Consequently, the proposed model is validated as a practical and promising model for forecasting nuclear energy consumption. (c) 2021 Elsevier Ltd. All rights reserved.

Automated summary

An optimized structure adaptive grey model is proposed in this paper, providing a generalized time response function theoretically and accurately modifying the background value based on Simpson's rule. Monte-Carlo Simulation and probability density analysis (PDA) reveal the robustness and reliability of the model. Experimental results show that the new technique significantly outperforms competitors from two different perspectives of PDA and level accuracy, and can still provide accurate forecasts across different forecast horizons.