Font Size: a A A

The Application Study Of Multivariable Predictive Control Method

Posted on:2016-10-17Degree:MasterType:Thesis
Country:ChinaCandidate:H SongFull Text:PDF
GTID:2348330542975741Subject:Control engineering
Abstract/Summary:PDF Full Text Request
In order to adapt to the modern industrial large scale systems optimization control requirements,the current widely used to control industrial process is predictive control technology as the core of the hierarchical structure.With the development of industrial technology,the main stream of prediction control technology adopts two-layer structure contains “steady-state optimization and dynamic control”.Two-layer structure prediction control can effectively solve the complicated industrial process common multi-objective optimization,multivariable control problem.Due to the development time is not long,compared with the industrial application,theoretical research is slightly inferior.In recent years,the algorithm design and the algorithm research on two-layer structure prediction control,the steady-state optimization problems such as feasibility analysis has certain research results.This subject on the basis of existing research,analyzed the structure characteristics and mechanism of two-layer structure prediction control,and combined with practical application,complete the rectification tower control problem.They subject a system software,to realize the related functional.The subject research process is divided into two stages,previous preparation and later data processing,the defect analysis.In the preparation process,get the step response model by system identification,and dynamic matrix control algorithm in conformity with the model information is selected as the control method,combined with two-layer structure prediction control mechanism in the steady-state optimization complete feasibility analysis of steady-state target layer,as a dynamic control layer set point,to perfect the relevant parameters and dynamic matrix control performance index information.Post treatment,the main is to use the existing models and existing data,and combining with the control characteristic of dynamic matrix control technology,to complete the control process of the system.The effect analysis is mainly to modify weight information about the control parameter,and simulate control process,come to a conclusion.Through modifying the weight parameters of different parameters,impact on system dynamic performance and steady-state performance.According to the conclusion,combining the actual requirement of industrial processes,complete parameters manually set make better control results tend to be the desired effect.Finally,through the software developed to realize the control process.Through the system control can make the control input and accused of output reach a steady state,change in the value of the permissible error scope.That implements the control function.The experimental results show that: this subject designed a software which can achieve the desired effect,complete industrial process control,and through the simulation process,summarizes the various parameters change basis weight coefficient.At the same time,this software is not order to achieve a single distillation tower control process research and development,but through the research process,aim to introduce more industrial control model,to complete the corresponding control.This subject has a considerable development,has practical significance.
Keywords/Search Tags:predictive control, two-layered structure, steady-state optimization, dynamic matrix control, performance index
PDF Full Text Request
Related items