A novel structure adaptive fractional discrete grey forecasting model and its application in China's crude oil production prediction. (30th November 2022)

• Record Type: Journal Article
• Authors: Wang, Yong; Ye, Lingling; Yang, Zhongsen; Ma, Xin; Wu, Wenqing; Wang, Li; He, Xinbo; Zhang, Lei; Zhang, Yuyang; Zhou, Ying; Luo, Yongxian
• Is Part Of: Expert systems with applications. Volume 207(2022)
• Publication Date: 2022-11-30
• Subjects: Structural adaptive -- Discrete grey model -- Fractional order accumulation operation with a parameter -- Monte Carlo simulation -- Probability density; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2022.118104
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: A novel structure adaptive fractional discrete grey forecasting model is proposed. A novel fractional accumulation operator with a parameter is developed. This paper conducts the robustness analysis of the FDGM ( 1, 1, k, r ) model. The prediction performance of the new model is better than the other six models. Crude oil production of the three regions in the next three years are forecasted. Abstract: Crude oil resources are related to all aspects of people's life, and play a vital role in the development of the national economy. Using nonlinear discrete data to reasonably predict crude oil production can help the government adjust energy structure and formulate energy development strategy, which has great practical significance. In this paper, a novel r-order accumulation operation with a parameter is proposed, and a novel structure adaptive fractional discrete grey forecasting model is established. Several classical optimization algorithms are compared, and the Grey Wolf Optimizer (GWO) is selected to calculate the parameters. For testifying the effectiveness of the model, a prediction model is constructed based on the total crude oil production in Qinghai, Liaoning and Shaanxi provinces of China, and a performance comparison experiment is designed with the existing six grey models. In addition, Monte Carlo simulation and probability density analysis provide a new perspective to further illustrate the robustness and accuracy of the proposed model. The experimental results show that this model is superior to the other six models in terms of fitting accuracy, prediction accuracy and model stability.