Performance Assessment、Oscillation Diagnosis And Optimization Of PID Control System In Thermal Process

PID controllers are widely used in thermal process.The performance directly influences the quality of regulation and the stability in unit’s operating.PID controller is designed according to a certain performance benchmark,and its performance gets worse as time passes by,which leads to the failure of reaching the desired performance target,resulting in a large gap between the actual performance and the desired requiements.What’s more,plenty of the control loops in the industrial process are in an oscillating state,among them,part of the oscillations are caused by the stiction of the control valve.Therefore,besides the performance assessment of the PID controller,diagnosis and compensation of the valve stiction are of practical use.This paper combines the following two aspects together:the deterministic performance assessment and optimization of PID control systems,and the main contents contain the following parts:1.In this paper,a comprehensive index has been proposed in terms of the load disturbance rejection and set-point following capabilities,as well as sluggish control detection of the PID control system.Firstly,a solution is developed for the calculation of the best achievable ITAE to preliminarily determine the perforamance of load disturbance rejection and set-point following.Secondly,the Idle index is put forward to judge whether the control loop is sluggish or not.Finally,the ITAE and Idle index are combined together and the ideal process time constantτ_c is analyzed for the selection of controller parameter.2.Control loop oscillation and valve stiction detection methods are researched.First,the detection method of loop oscillation is proposed,which is performed by the time domain method.Then,the area index is added to judge the strength of oscillation.Second,on the basis of oscillation detection,taken the Kano two-parameter model as the stiction model,a simulation loop is then set up to obtain the corresponding simulation data and calculate the cross-correlation function of the controller output and the process output.By analyzing the parity of the cross-correlation function,the stiction characteristic is qualitatively judged.When the sticiton is small,it can be optimized by adjusting the controller parameters without valve maintenance.Thus,the strength of stiction needs to be judged on the result of the stiction detection:calculation of the size of the stiction interval and the deviation between the controller output and the valve output.This delamination-detection method can not only qualitatively describe the viscous properties but also quantitatively calculate the size of viscous parameters.3.A solution to optimize the PID control system is studied and proposed.On the basis of the quantitative calculation of the valve stiction characteristics,a series of short pulse signal are added to the manipulated variables to eliminate the friction of the valve.In addition,if there no valve stiction detected,the perforamce of the control system can be improved by parameter optimization.And the ideal process time constantτ_c can be determined by the combination of ITAE and Idle index,then the parametric model of the reference controller can be obtained.