Jinfang Zhang

Adaptive Neuro-control System for Superheated Steam Temperature of Power Plant over Wide Range Operation

In this paper, the dynamics of a superheated steam temperature process in a 600MW supercritical once-through boiler is first analyzed. The dynamics is influenced by the operating conditions represented by steam flow and steam pressure mainly. An intelligent cascade control system is then designed for controlling superheated steam temperature over wide range conditions. The parameter of the proportional controller in the inner loop is fixed all the time. The controller in the main loop is a self-tuning PID neuro-controller whose parameters are adjusted by gradient algorithm, where the Jacobian information of the plant is obtained by RBF neural networks. Finally, the designed control system is applied to control the superheated steam temperature in a 600MW supercritical once-through boiler, and the simulation indicates it has satisfactory performances over wide range

Generalized predictive control for wind turbine systems

It is difficult to deal with the variable speed constant frequency (VSCF) wind turbine systems due to the stochastic characteristics and the pneumatic effects of wind. In this paper, a new pitch controller based on generalized predictive control theory is designed to improve the power-output quality of variable speed constant frequency wind turbines. An application to a 300MW wind turbines is given, and simulation results show that the proposed method is effective in wind speed interference suppression and constant out power.

A rational square-root B-spline model approximation and control of output probability density functions for dynamic stochastic systems

Based on the analysis of the relationship between the actual weight and the pseudo-weight of the rational square-root B-spline based PDF model, this paper presents a new method to the modeling of the rational square-root B-spline based PDF model and the controller design is discussed based on the model. Both the modeling and controller design algorithms are applied to the simulation study of the molecular weight distribution (MWD) process, the results verify the validity of the algorithms.

Probability density function predictive control based on the rational square-root B-spline model

A predictive control algorithm for probability density function (PDF) is presented based on the rational square-root B-spline based PDF model. The rational square-root PDF model is studied first, based on the analysis of the relationship between the pseudo weight and the actual weight, the rational square-root B-spline model can be easily set up; then the predictive model is deduced; a performance index is chosen next to design the predictive controller; and simulation study is performed for the molecular weight distribution (MWD) control of a styrene polymerization process to test the validity of the predictive control algorithm.

Minimum entropy incremental control for nonlinear stochastic systems with non-Gaussian disturbances and uncertain parameters

In this paper, the problem of controller design for a general nonlinear stochastic system is investigated. The randomness of the considered stochastic system comes from external excitations and system parameters. To achieve minimum entropy control, the probability density function (PDF) of the tracking error is given by using recursive method firstly. Then, a new performance index is formulated, which contain the entropy of the tracking error, the mathematical expectation of the squared error and the control energy. A recursive optimization control algorithm is obtained by minimizing the performance index. Stability analysis is established and encouraging simulation results are included to illustrate the efficiency of the proposed method.

Improvement of boiler combustion control performance using probability density function shaping and particle swarm optimization

A novel scheme for boiler combustion control system is proposed, where the firing rate demand is represented by furnace temperature distribution (FTD). In this scheme, the firing rate is defined to be FTD which characterizes the status of combustion directly. A real-time FTD shape control system is formulated and applied to boiler combustion control. FTD is fed back to combustion controller and adjusted by fuel valve actuators. Compared with the gradient approach, particle swarm optimization (PSO) is used to solve the control input, the simulation results verify the effectiveness of the proposed scheme.

Networked Control For Robotic Manipulators Via T-S Fuzzy Control And Descriptor Representation

This paper presents a novel networked control scheme for robotic manipulators by combing T-S fuzzy control with descriptor representation. The dynamic characteristic of nonlinear control system for robotic manipulator is built by T-S fuzzy model. In specific, both time-delay induced by network and packet loss are considered. A delay-dependent criterion for determining the stability of networked control system is obtained with the aid of linear matrix inequality (LMI) approach. A fuzzy controller is then designed by parallel distributed compensation principle. An augmented system is also introduced such that the tracking error is minimized. Finally, the proposed control algorithm is applied in a networked single-link robotic manipulator control system, and the result of simulation shows its effectiveness

Stochastic distribution control for the permanent magnet synchronous generator based variable speed wind energy conversion systems

In this paper, the nonlinear dynamic model of the wind energy conversion system (WECS) is established. Due to the non-Gaussian randomness in the WECS introduced by the wind speed, stochastic distribution controller is designed for both maximum power point tracking (MPPT) and attenuation of disturbance. Compared with conventional PI controller, the effectiveness of the proposed approach is verified by some experimental results.

Boiler steam temperature control based on Linear Active Disturbance Rejection Control

Boiler steam temperature in thermal power plant is very complex and hard to control, many control strategies have been applied to it. Linear Active Disturbance Rejection Controller (LADRC) is adopted to the control of boiler steam temperature in this paper. The nonlinear model of boiler steam temperature is set up firstly, then the principle of the LADRC is briefly introduced and the LADRC based boiler steam temperature closed loop control system is designed and the closed loop system is studied with simulation. Compared with the traditional PID, more robust and better disturbance rejection properties are obtained.

Combustion Control using B-Spline Based Furnace Temperature Distribution

This paper presents a closed-loop combustion control algorithm for a coal-fired boiler using the recently developed output distribution control theory. The furnace temperature distribution (FTD) is regarded as the controlled variable, which demonstrates the stability, economy and reliability of combustion process. Following the FTD obtained by Fluent simulation software, the equivalent B-spline FTD model is then established. Consequently, the control strategy is provided so as to ensure the output FTD will follow the desired distribution. The simulation study shows the effectiveness of the proposed approach.