Font Size: a A A

Based On Liquid Level Predictive Control For More Warter Tank Of Wavelet Neural Network

Posted on:2018-03-21Degree:MasterType:Thesis
Country:ChinaCandidate:K JiangFull Text:PDF
GTID:2348330512966979Subject:Control engineering
Abstract/Summary:PDF Full Text Request
As the increasing demand for high-precision products,today's increasing demand is prompting the continuous progress of modern science and technology.Especially in the field of intelligent control,the original or traditional algorithms have been very difficult to meet the needs of today's society.So the need for continuous innovation in technology,for intelligent control,that is to improve the algorithm.Many industrial sites are extremely complex and it is important to have a theoretical preparation and experimental work to simulate a complex site in order to achieve a better control strategy.The liquid level control system of the four-tank is a good experimental system,and the system has certain non-linear coupling relationship.It can decouple the water level of two adjacent water tanks,but also can control the level of four water tanks by multivariable.Therefore,for nonlinear and coupled control,it is a good experimental object,you can predict control to meet its control requirements.Therefore,in this paper,the relationship between the nonlinear and coupling to the tank level as the controlled object,launched the design.Wavelet neural network is known for its good non-linear approximation ability,which determines the accuracy of the model.Therefore,the parameters of wavelet neural network are optimized firstly.Optimization of Wavelet Network Based on Improved Particle Swarm Optimization.Particle Swarm Optimization(PSO)algorithm introduces the genetic crossover factor,the linear decreasing weight and the average individual extremum.Which greatly improves the approximation ability of wavelet neural network.And then combined with the model algorithm control strategy for the four-tank water level control.In order to solve the coupling problem of water tank,neural network decoupling technology is introduced.And the related parameters of the decoupler are designed.The results of simulation show that the control algorithm has good controlperformance,high precision,fast response speed,strong robustness,and also has certain anti-jamming capability.The simulation results show that the control algorithm has good control performance,high precision,fast response speed and robustness.The experiment in this paper provides a better solution to solve the problem of nonlinear coupled system.
Keywords/Search Tags:wavelet neural network, improved particle swarm optimization, model algorithm control, decoupling technique
PDF Full Text Request
Related items