Font Size: a A A

Research On Optimization Design And Parameter Self-Tuning Of Controllers For Plants With Uncertainties

Posted on:2007-02-23Degree:MasterType:Thesis
Country:ChinaCandidate:Y P DuFull Text:PDF
GTID:2178360182970816Subject:Pattern Recognition and Intelligent Systems
Abstract/Summary:PDF Full Text Request
Designing control system for plants with uncertainties is one of the forward positions of control theory. To the plants with uncertainties, several control methods with self-tuning parameters and PSO algorithm based optimal design for the controllers are proposed in this thesis. The experiments of basis weight control, controlling hydraulic turbine generators are made. The simulation results show that the proposed controllers have good performances. The main contents of this thesis are as follows:1. The PID control method with neuron tuning parameters and the PID control method with PSO-neuron tuning parameters are proposed for multi-model plants in this paper. The initial values of the former controller can be fetched without limit, the initial values of the later one are determined by a PSO algorithm. The PID controller is designed to control a multi-model plant, and the neuron is used to regulate the parameters of the PID controller on line. With an example of the basis weight control, the simulation experiments are made. The results illustrate that the proposed controllers are available to multi-model plants.2. To the plants with grave uncertainties, the neuron control method with fuzzy tuning parameters is proposed in this paper. In this control system, the neuron is designed to produce the control signal in model-free way, and the neuron controller parameters are tuned by fuzzy algorithms. With examples of hydraulic turbine generators, the experiments are made. The simulation results show that the proposed controller has good performances under the operating conditions of various guide vane opening, fast transient response and strong robustness.3. Based on the particle swarm optimization (PSO) algorithm, the optimum design methods on-line or off-line for neuron controllers are proposed in this paper. The proposed methods are used to design the optimum neuron controllers for hydraulic turbine generators. The simulation tests are made. The results show that satisfactory performances are reached.
Keywords/Search Tags:neuron control, plants with uncertainty, self-tuning parameters, particle swarm optimization (PSO) algorithm, Optimization
PDF Full Text Request
Related items