Soft Measurement Of Temperature In Vacuum Sintering Furnace Based On Mechanism And Data-Driven Hybrid Modeling

At present,the temperature sensor cannot directly measure the internal high temperature of the vacuum sintering furnace,so it is impossible to accurately control the temperature of the vacuum sintering furnace in actual production.And vacuum sintering furnace temperature change is crucial to the character of the sintering material,so the internal temperature of vacuum sintering furnace for precise control,can not only in the quality of sintered products have safeguard,also can control the uncertainty factors in the process of firing and adjust in time,in order to ensure the normal operation of the sintering process,aimed at the problem in this paper,based on the mechanism and data driven hybrid modeling method of soft measurement of the temperature of vacuum sintering furnace is studiedWith the development of soft measurement technology and engineering application is more and more widely,its main way is through mechanism analysis,a study on soft measurement data driven,and only with mechanism analysis alone or with a data-driven method,unable to accurately describe the vacuum sintering furnace heating process of nonlinear and strong lag problem,temperature of vacuum sintering furnace is difficult to change the real-time online prediction.Aiming at this situation,combining with the mechanism modeling method to establish the temperature model,on the basis of using the measured experimental data and simulation data for the temperature error compensation,using mechanism analysis and data driven hybrid modeling method to make up for the deficiency of the two,makes the temperature of the model is set up close to the actual conditions,make the model more convincingThe main content of this thesis is to soft-measure the temperature of vacuum sintering furnace.The main contents are as follows:mechanism temperature modeling,numerical simulation,temperature error compensation based on BP neural network,realization of temperature prediction system of vacuum sintering furnace.Firstly,by establishing the mechanism model of vacuum sintering furnace temperature,the electro-thermal conversion model and the mechanism model between the heating body and the workpiece in the furnace can be obtained.Then,according to the 3d model of vacuum sintering furnace and the actual process parameters,the internal temperature field was simulated by ANSYS-fluent finite element analysis software to obtain the temperature data.The BP neural network is used to study the temperature error compensation of the existing experimental data and the simulated temperature data on the basis of the mechanism modeling.Finally,the temperature prediction system of vacuum sintering furnace is established.The specific work is as follows(1)In order to solve the problem that the temperature in the vacuum sintering furnace cannot be accurately grasped in the working process,the method of soft measurement is used to study the temperature.True mechanism of sintering furnace temperature model is established first,analyzing the physical process of heating up,the temperature model is set up,for the structure characteristics of vacuum sintering furnace,artifacts,distribution,working condition of thermocouple,and the distribution of the heating pipe in detail,in order to facilitate the subsequent chapters in the numerical simulation analysis of temperature field distribution of the furnace.At the same time,the radiant force,system emittance and Angle coefficient of the vacuum sintering furnace in the process of radiant heat transfer were calculated by combining the working principle of the furnace and the heat transfer process analysis(2)In order to solve the high temperature circumstances,vacuum sintering furnace internal parameters can’t directly access problem,according to the working condition of the temperature of the model parameters of choosing materials,and work environment setting itself as the research object,using the finite element analysis software to simulate the temperature field simulation of vacuum sintering furnace,heating process under different process parameters set for simulation,analysis and the simulation results with the experimental data acquisition analysis and comparative test and verify the reliability of the simulation results,acquire the temperature data of research at the same time(3)In order to solve the problem that the modeling of a single mechanism is not accurate,the BP neural network is used to compensate the temperature error on the basis of the mechanism modeling,so that the established temperature model is closer to the furnace temperature under the actual working condition.In this part,BP neural network is used to calculate the temperature error value by combining the experimental data and simulation data.Finally,the temperature model under the mixing model is obtained,and the temperature difference between the model under the mixing model and the model under the single mechanism is compared with the experimental data(4)According to the temperature model established above,MATLAB GUI software was used to design the temperature prediction system of vacuum sintering furnace,and the functions of each module were established to control and adjust the temperature of vacuum sintering furnace in real time.