Performance Assessment And Improvement Of Control System

The control engineers have been committed to doing research on the theory, design and application of control system. The purpose of designing a control system is to maximize the profit from raw material to products while meeting various constraints, such as product quality constraints, operation constraints, safety constraints and so on. To obtain the maximum benefits, control system is required to be in good performance all the time. In the early stage, the control system can be in good performance, but after a long run the performance of the system will deteriorate owing to change of material and product characteristic. Performance deterioration is disadvantage to maximum benefits. So it is necessary to monitor the performance of the system.The main work and achievements in this paper consist of the following aspects.First, feedforward/feedback control system integrates the advantages of both feedback control and feedforward control and is widely used in industrial field. So this paper studied the performance assessment method of feedforward/feedback control system, and took the boil water supply system as an example for simulation. Simulation results show that when measurement noise exists in the system, adding feedforward controller can improve the performance of the system. However, the feedforward controller must be designed properly, otherwise it will deteriorate the performance of the system.Second, for complex control system, cascade control is the widely used control strategy. So on the basis of previous work, this paper further studied the performance assessment method of cascade control system, and took the main steam temperature control system as an example for simulation. On one hand, this paper combined the generalized minimum variance benchmark with the minimum variance benchmark to assess the main steam temperature control system. Assessment results show that the combination of the two benchmarks can fully reflect the performance of the system, and by comparing the two benchmarks can find the reason of poor performance. On the other hand, because of the characteristics of large delay of main steam temperature system, this paper found the strategy which can improve the performance more effectively by comparing the performance of different control strategies. Simulation results show that the performance of Smith ADRC is better than other control strategies.Third, for complex multivariable control system, this paper first put forward a set of practical performance assessment method based on the upper and lower bounds of the minimum variance and took the coordinated control system in power plant for simulation. This paper found the strategy which can improve the performance more effectively by comparing the performance of different control strategies. Simulation results show that decoupling ADRC control can further remove the coupling between variables, thus improving the performance of the system.The last, considering there are a lot of nonlinear systems exist in the industrial field. So this paper studied the performance assessment method of nonlinear system. Two performance assessment methods were summarized in this paper, removing the nonlinear term method and nonlinear modeling method. The paper mainly studied the first method, found reasonable nonlinearity extraction and nonlinearity test method, and then made a simulation using these methods. Simulation results prove the rationality of these methods.