Research on building material design in circular economy based on improved neural network.

Material management technology has always been the basic and core technology of traditional construction projects. But now, because traditional material management consumes more resources and has low efficiency, it has gradually been replaced by modern material management systems represented by computer technology, thus saving a lot of enterprise input costs and reducing the impact of traditional material management technology on enterprise operating income. Based on this background, this paper introduces improved neural network technology to adapt to the circular economy environment, so as to complete the construction and improvement of the building materials data management system. The system can be used to complete building materials management, basic data management, etc., and can be optimized for the system itself. Through network technology, the client can be connected to the server switch, and Ajax technology and JSP programming language can be used for compatibility processing, so that the system can run better in the Windows environment. According to the simulation experiment designed for the system, the output sample value of the system is close to the measured value, with good prediction accuracy, small error and high stability. Therefore, the application of AdaBoost ANN algorithm can effectively realize the prediction of building materials. In addition, other functions of the system, such as interactive page design, operation interface and database, can better adapt to the circular economy environment and actual user needs. In this paper, the improved neural network is applied to the field of building material design under the environment of circular economy, thus completing the system optimization. [ABSTRACT FROM AUTHOR]