Energy Efficiency Evaluation Model of Prefabricated Buildings Based on Graph Convolutional Neural Network
- Resource Type
- Conference
- Authors
- Wu, Xiang; Qi, Dianwei
- Source
- 2023 IEEE 7th Information Technology and Mechatronics Engineering Conference (ITOEC) Information Technology and Mechatronics Engineering Conference (ITOEC), 2023 IEEE 7th. 7:648-651 Sep, 2023
- Subject
- Communication, Networking and Broadcast Technologies
Computing and Processing
Robotics and Control Systems
Signal Processing and Analysis
Surveys
Prefabricated construction
Technological innovation
Mechatronics
Pandemics
Green buildings
Convolutional neural networks
graph convolutional network
prefabricated construction
energy-saving building
accuracy
- Language
- ISSN
- 2693-289X
The evaluation method of graph convolutional neural network has been developed and perfected after a long time of development. The advantage of graph convolutional neural network is that its highly simulated neural operation function can effectively correct errors after repeated learning, so as to avoid the influence of subjective factors on the results to a certain extent. At the same time, the graph convolutional neural network is based on nonlinear operation, which is different from the traditional calculation method which mainly relies on carding regression. It can better grasp the relationship between variables, and the results of hidden variables in the calculation can be directly fed back to the calculation results.