Active Disturbance Rejection Controller And Its Application Research

On the basis of a long-time and deep research on PID and modern control theory, the Auto disturbance rejection control (ADRC) was invented by Han Jingqing, a researcher of System Science School of China academy of sciences. The controller has many merits such as simile arithmetic, easy-realized, highly precise, highly anti-interference, and so on. This paper mainly studies the application of ADRC algorithm in the higher-order, nonlinear and coupling systems. The detail studies as follows:1. This paper presents a method to replace the practical plant characteristics with low-order time varying difference equation for a constant system of parameters unknown and high order. The cascade extended states observer (CESO) is used to estimate the states of continuous-time system and the total disturbances, and then the least-squares algorithm with forgetting factor is applied to estimate the system parameters. In steady state, the characteristic model is equivalent to the output of the practical plant, so the controller can be designed according to the characteristic model and compensation from the disturbance estimated by the CESO.2. For the problem of coupling in a class of multiple input multiple outputs (MIMO) systems, this paper proposes a dynamic nonlinear decoupling control method based on the ADRC. This method does not rely on the accurate model of the system. According to the partly known and unknown coupling matrix of the control part respectively, this method takes the model perturbation, the external disturbance and the dynamic coupling including the interaction of the input variables as a total disturbance to each channel based on the local static decoupling. By introducing virtual control and state variables, the extended state observer (ESO) is designed to estimate the total disturbance and then it is fed back to the controller to compensate the disturbances. Then for each decoupling sub-object, the non-linear SISO ADRC is designed to ensure the stability of the closed-loop system.3. Arming at the problem of over many parameters to be turned in the ADRC, a self-adaptive single neuron Active Disturbance Rejection Control (SNADRC) is proposed based on RBF neural network on-line identification, which identifies the Jacobian matrix of controlled object by means of RBF neural network identifier and acquires on-line tuning information of ADRC parameters. The self-tuning of the controller parameters is implemented by the single neuron controller, and the intelligent control of the system is achieved.4. In light of the large delay, strong inertia, and uncertain dynamic characteristics of Continuous Stirred Tank Reactor (CSTR) system, a discrete active disturbance rejection control scheme is proposed, and then the latest SIEMENS programmable controller of SIMATIC PCS7 is employed in the emulation CSTR object which fabricated in the new multifunction process and control experiment system to gain stable control of temperature. The result of the tests shows that the main steam temperature control system with ADRC ensures very good robustness and adaptability under modeling uncertainty and external disturbance, and produces better dynamic performance than classic PID controller in the operation conditions.