Intelligent Detection System For Vacuum Prediction Of Vacuum Glass

As a new generation of energy-saving environmental protection materials,vacuum glass has insulation,anti-condensation frost,sound insulation and other excellent performance.The performance of vacuum glass can be characterized by the degree of vacuum,the higher the vacuum,the better the performance of vacuum glass.However,studies of vacuum glass properties over the years have shown that actual measurement of vacuum glass performance characterization is difficult.And with the accumulation of time,the indicator vacuum degree of the vacuum glass is also declining,making the vacuum degree in the vacuum glass measurement becomes a high difficult subject.Aiming at the current methods of vacuum glass performance testing,we have not adopted the intelligent model based on the intelligent algorithm to establish various possible factors and the performance parameters of vacuum glass.In this paper,based on the heat transfer process of vacuum glass,an intelligent detection system for vacuum glass vacuum prediction is designed.The intelligent detection system mainly includes wireless temperature detection,intelligent algorithm model and the host computer management software three parts.Based on STM32F103 microcontrollers,MLX90614 infrared temperature measurement module and nRF24L01 radio frequency module,this paper designed a set of wireless temperature detection system to detect the temperature data in the experiment,and use the USB interface to transfer the temperature data to the computer side for related processing.Designing the experiment,to heat the side of the vacuum glass(heat source side or hot plate)at constant temperature,the temperature data of the other side of the vacuum glass(non-heat source side or cold plate)and the room temperature data of each time period are measured in a period of time.Using MATLAB software as the platform,BP neural network algorithm is used to establish the intelligent model,so as to predict the vacuum degree of the vacuum glass.The prediction result is ideal.In addition,using Visual Studio 2015 as a platform,based on the WPF design a host computer management software,choosing SQL Server database to storage data,using MATLAB extension function MATLAB Builder for.NET to package BP neural network algorithm the establishment of the intelligent model into a form for the.NET host computer application to use.