Gain self-scheduling pid controller for dual-input and double-output systems in thermal power plants

Abstract

The invention provides a self-gain-scheduling PID (proportion integration differentiation) controller of a double-in double-out system of a heat-engine plant. Through a quick / slow loop detection setter, a quick / slow loop control mode switch, quick and slow loop feedforward controllers and quick and slow loop feedforward-feedback comprehensive arithmetic units, and through lap joint of proper modules, relay feedback control of a general double-in double-out system loop is realized, critical vibration information of the process is measured, and under the condition that a controlled system mathematic model is not known, P, I and D parameters of the double-in double-out system loop can be automatically set in closed-loop control according to stability margin as required; according to changes of actually-measured operation condition signals and deviation signals, optimal P, I and D parameters of quick and slow loops are automatically inferred in real time, transition time of a control system is reduced to the utmost, and dynamic and static deviations of the control system are reduced. By the self-gain-scheduling PID controller, parameter setting of the multivariable control system becomes more quick and scientific, and the control system has high robustness.

Description

technical field [0001] The invention relates to the technical field of multivariable control of process production process, in particular to a gain self-scheduling PID controller of a dual-input and double-output system in a thermal power plant. Background technique [0002] In a thermal power plant, it is a multi-variable and complex production process, and there is a strong correlation between the variables. For this, a generalized double-input and double-outlet control strategy must be adopted, while changes in the operating conditions of the unit and fuel supply, etc. , such as load changes and coal quality changes, will cause relatively large changes in the dynamic characteristics of production control objects, making this type of control objects show strong coupling, time-varying, and nonlinear. However, modern multivariable control methods require a complete process model, which cannot be obtained accurately in many cases, making many multivariable control methods dif...

AI Technical Summary

Problems solved by technology

In fact, the production process of power plants is time-varying and non-linear. When the load, coal quality, etc. change, the object characteristics will be serious. If the controller parameters are not changed, the control quality cannot be guaranteed.

In the past, the parameter setting of each PID controller was preset by experience, and then manually adjusted according to the response waveform of the operation. The differential effect is not easy to adjust. The differential link is often removed, and the effect of differential predictive control cannot be achieved.

The PID control parameter setting of the double-input and double-output system is a time-consuming and laborious work.

Images