In this branch, we demonstrate that the joined model of Kernel Principal Component Analysis (KPCA) and Linear Regression (LR) outperforms other plans, thereby offering a novel and doable approach for long-term power consumption guess. We propose a linked model of KPCA and LR for this purpose. Despite a limited sample intensity, the model can accurately forecast the momentary changes in total electricity consumption, bragging high interpretability and useful utility. We working KPCA to reduce the complicatedness of the original data, before input the dimensionally shortened data into a Backpropagation Neural Network (BPNN) and other models, flexible optimal model results. Visualization of the three principal elements derived through KPCA disclosed that the first principal component represents the unending growth of power consumption, while the other two parts represent the unending fluctuation of power consumption. Additionally, people features, price visage, and industrial building features also cause the increase in China's electricity use, albeit in a changeable manner. Lastly, we foresee that China's total societal power consumption will reach 1.83 heap KWH by 2035, a forecast more optimistic than that of Oxford experts and regular with China's success in combating COVID-19. The model holds an positive view of China's future economic prospects, joining with China's brisk economic growth and allure comprehensive win in the fight against COVID-19.
Author(s) Details:
Zili Huang,
School of Science and Engineering, The Chinese
University of Hong Kong, Shenzhen, Guangdong-518172, China.
Haochen
Zhang,
School
of Aerospace Engineering, Tsinghua University, Beijing-100084, China.
Chenxi Qiu,
School of Science and Engineering, The Chinese University of Hong Kong,
Shenzhen, Guangdong-518172, China.
Jia Liu,
Renm Consulting Company, Beijing-100038, China.
Please see the link here: https://stm.bookpi.org/FRAPS-V6/article/view/10834
No comments:
Post a Comment