Research On Intelligent Modeling Of The Heat Transfer Process For Vacuum Glass Based On Soft Sensor

The most important parameter representing the thermal properties of Vacuum Glass(VG),the heat transfer coefficient,is difficult to measure online because it increases over time,thereby decreasing thermal insulation performance.Determining the heat transfer of vacuum requires detailed knowledge of the thermal properties of their different elements.A series of standards and guidelines exist in this area.The thermal properties of the frame can be determined either by detailed two-dimensional numerical methods or by measurements in accordance to European or international standards.First,the study concerning intelligent modelling of the mechanism heat transfer vacuum glass based on Soft Sensor to get their vacuum glass performance data is proposed.This method ensures that the feasibility of intelligent modelling and provides a theoretical basis for predicting performance parameters of vacuum glass insulation to intelligent modelling based on Soft Sensor.The study was conducted to develop an efficient method to simulate heat transfer through vacuum glass.Based on advanced numerical simulation technology,computational fluid dynamics software was used to analyse the heat transfer process,and the simulation results applied to guide and analyse the nonsteady-state test method.This approach guarantees that the center of the heating plate undergoes onedimensional heat transfer,and the temperature measurement at the center of the non-heated surface is of practical significance,it is necessary for the study intelligent modelling and prediction of thermal insulation performance.Second,we applied neural network methods to predict the heat transfer coefficients of vacuum glass.Based on MATLAB software,a neural network intelligence model was established,and the traditional back-propagation(BP)neural network was optimised.A genetic algorithm was used to reduce the dimensions of the independent variable.Then,the Mind Evolutionary Computation algorithm was used to optimise the initial weight and threshold.Using the optimised BP neural network intelligence model to predict the heat transfer coefficient of vacuum glass insulation,we derived an average absolute error of 0.0076.Third,an online prediction model for thermal insulation of the vacuum glass based on an optimized Least Squares Support Vector Machine(LSSVM)is proposed.Numerical simulation experiments verify that the RBF(Radial Base Function)neural network and the LSSVM technology the intelligent model can monitor the central temperature of the glass under vacuum without a heat source.Based on the RBF and LSSVM neural network,predicting the lateral center temperature not linked to the heat source of the vacuum glass is proposed to guarantee the model's prediction accuracy,it's effective with very high accuracy.At last,we have also studied new ways of predicting the lifetime of vacuum glass.The purpose of this method is to predict the lifetime of vacuum glass based on Fuzzy.This method absorbs the ambiguity and randomness of the observed values based on the randomness of prediction technology.It makes the forecasts more realistic and more suited to the analysis of the life of the vacuum glass.Therefore,this method has great practicality.