Study On Control Performance Assessment Based On Fuzzy Models

It has been widely recognized that how to enhance industrial equipment availability and reliability of devices is important to improve the operation of the production process reliability,safety and economy.Because the process is inevitable that there are some kinds of device changes or other kinds of disturbances,the industrial sector has been generally recognized that the use of the process of measured data on the performance of control systems online or offline evaluation and diagnosis,can be early detection and diagnosis of process equipment and process,control equipment and the abnormal state of failure,and realize an effective control system maintenance,optimization of control system parameters,therefore the real-time control system performance assessment is to improve system availability and security to run an effective measure.Control performance is mainly affected by sensors,the executing agency (such as valves) and the device performance and the impact of disturbance. Work equipment anomalies or failures,often the introduction of non-linear control loop,resulting in the control of linear systems based on performance evaluation methods are no longer applicable.Minimum variance control principle is an effective and feasible principle, and also controller performance assessment employing.Only routine closed loop operating data and time delay performance of process are needed making use of minimum variance principle as a benchmark in controller performance assessment.The paper present control loop performance assessment methodology based on fuzzy model identification mainly for a class of nonlinear industrial processes.