Peak emission path of rural carbon emission in Chongqing based on optimization neural network and genetic algorithm

作     者：Quan, Feng Liu, Peng Quan, Kai Zhang, Xingxing Jiang, Wenjie 

作者机构：Chongqing Water Resources and Electric Engineering College Chongqing Yongchuan China Art and Media College Chongqing Metropolitan College of Science and Technology Chongqing Yongchuan China  Bangkok Thailand Science Multimedia Journalism School of Multimedia Technology & Communication Awang Had Salleh Graduate School of Arts and Sciences Universiti Utara Malaysia Kedah Sintok Malaysia  School of Computing Awang Had Salleh Graduate School of Arts and Sciences Universiti Utara Malaysia Kedah Sintok Malaysia College of Landscape Architecture and Arts Northwest A&F University Shaanxi Yangling China 

出 版 物：《Journal of Biotech Research》 (J. Biotech Res.)

年 卷 期：2024年第19卷

页      面：269-280页

核心收录：

基　　金：This work was sponsored by Institutional Research Projects at Chongqing Water Resources and Electric Engineering College in 2024 (Grant No. K202401) and Scientific and Technological Research Program of Chongqing Municipal Education Commission in 2022 (Grant No. KJQN202203805) and 2023 (Grant No. KJQN202302502) 

摘      要：The growth of industrialized agriculture has caused the problem of excessive carbon emission (CE) in animal husbandry and rural agriculture. To maintain energy conservation and carbon reduction in rural economic development, this study explored the peak carbon emission path in Chongqing, China through the optimization model of back-propagation neural network and genetic algorithm. The research analyzed the optimization direction of energy structure through the carbon emission of rural industrial composition with the aim of projecting energy use efficiency, which was achieved by weighing the financial gains from energy use against the associated carbon emissions. This study extended the direction of industrial optimization by combining the cross-variance model of genetic algorithm to explore the global optimal path through population simulation. The study also improved the adaptability of the model by iterative training method with the help of the learning process of back propagation neural network adaptive data. The results revealed that the loss function of the proposed model basically converged after 30 iterations in the test set, and the prediction accuracy of the model could reach more than 80% after 60 iterations. The fitness value of the proposed model was reduced to 0.22 × 10-3 after 120 iterations, while the lowest fitness value of other algorithms could only be reduced to 0.38 × 10-3, which indicated that the optimization effect of the proposed model was significantly better than other methods and could effectively avoid the local optimal solution problem. The proposed model could provide an effective planning path for exploring the peak carbon emissions in the rural area of Chongqing, China. © (2024), (Bio Tech System). All rights reserved.