On this page

Deep Neural Network-based Dynamic Sensing and Emergency Response Technology for Environmental Protection Risks during the Construction Period of Transmission and Substation Projects

By: Xiaohu Sun 1, Xiaofeng Chen 1, Shu Zhu 1, Yanbing Wang 1, Qing Li 2, Zhengang Wang 3
1State Grid Economic and Technological Research Institute Co., Ltd., Beijing, 102209, China
2Hubei Anyuan Safety & Environmental Protection Technology Co., Ltd., Wuhan, Hubei, 430000, China
3State Grid Information and Telecommunication Group Co., Ltd., Beijing ,100069, China

Abstract

Due to the lack of control and technology in the development and construction of new power systems, the current construction of transmission and substation projects in certain regions still have environmental risks that cannot be ignored. Based on the characteristics of the power grid, this paper proposes eight power environmental assessment indicators. On the basis of the definition of the indicators, the calculation method of the indicators and the scoring are designed. At the same time, BIM technology and BP neural network algorithm are integrated to design the processing method of transmission and substation engineering data. Based on the Bayesian network algorithm, the steps of environmental risk assessment during the construction period of transmission and substation projects are explained, so as to establish the environmental risk assessment model. The expert scoring method and principal component analysis are adopted as the practical application methods of the assessment model, so as to realize the dynamic perception of environmental protection risk during the construction period of transmission and substation projects. The environmental risk assessment model constructed has a good consistency in assessing the probability of occurrence of risks at different stages of the construction period as 85.7%, 79.9%, and 89.7%, respectively, and the model is able to perceive the environmental risks of transmission and substation projects during the construction period more accurately.