Design Of Control System And Research On Temperature Control Methods For Tunnel Kiln

Tunnel kiln is an important heat equipment,it's widely used for refractory products production.The automatic control level and the temperature control method of tunnel kiln play a unique role in the quality of the product and energy consumption of production.Designing a reasonable tunnel kiln control system and developing an advanced,effective temperature control method are two of the research directions of tunnel kiln control field.The paper focuses on these two aspects.The paper takes the actual project as the background.After comprehending and analyzing the production process,process characteristics and control system's requirements of tunnel kiln,the control system of tunnel kiln is designed.It is consists of two parts,the basic control level function and the monitoring management level function.The basic control function is developed using Siemens PLC to achieve the loop control of temperature,pressure and flux.The monitoring management function is developed using Microsoft Visual studio 2012 platform and C#language to achieve the function of displaying the thermal parameters,production management,etc.After establishing the temperature model of tunnel kiln,it began to research the temperature control method.Owing to the tunnel kiln runs in a complex situation,it is hard to build a precise mathematical model.This paper uses CARMA(Controlled Auto-Regressive Moving Average)to describe the temperature controlled system approximately.According to the data collected in the field,the order of the temperature model is determined by the ratio determinant method,and the parameters of the model are identified by the gradient correction parameter identification method.From the simulation results,we can see that the temperature model established by this method is effective.Since the tunnel kiln system is a complex system with nonlinear,time-varying,and lagging characteristics,it is difficult to achieve the desired control effect by using the conventional PID control method.The PID controller based on BP neural network can realize the self-tuning of PID parameters.Its control performance is better than the traditional PID control.However,it is easy to fall into the local optimum for the BP neural network,and the initial weight selection has great influence on the control effect.The Jacobian information in the weight adjustment process adopts the approximate value,which has an affect to the accuracy.This paper puts forward measures to solve the above problems.Firstly,the improved PSO algorithm is used to optimize the initial weights of BP neural network.Then,the RBF neural network is added to the control loop to identify the Jacobian information of the controlled object online,which is used to adjust the weights.Simulation results show that,the following performance and anti-interference ability of the improved method are better than the method before improvement,and it still has a good control effect in the case of model mismatch.