Research On The Coordinated Control Of Ship Steam Power Unit Based On T-S Model

The coordinated control system is the core component of the ship’s steam power plant,the structure and thermal dynamic process are complex,and it has strong nonlinearity,large delay,multi-variable coupling and multi-constraints.With the continuous development of the automation level of the ship in recent years,the requirements of control performance and response performance are higher.Based on the analysis of the dynamic characteristics of the controller,multi-model modeling methods and predictive control strategies,this article focuses on modeling mathematical models and controller design.The main works as follows:(1)Analyze the composition,classification and characteristics of the control system.According to the control level,the coordinated control system is divided into three parts: the main control system,the local control system and the control object.According to the way of working,it is divided into two types: boiler follow and turbine follow.And this chapter briefly introduce the actual furnace unit.(2)The basic principles of the T-S fuzzy multi-model modeling method are introduced from four aspects.Focus on the computational burden will increase sharply while the number of clusters of the T-S fuzzy modeling method increases,an improved genetic algorithm to optimize parameters is proposed.The Box-Jenkins gas stove is simulated in MATLAB/Simulink,by comparing the output mean square error of the two models,the effectiveness of the improved modeling method in improving the accuracy of the model is verified.(3)This chapter is research on the application of the improved T-S fuzzy modeling method in the coordination system.First,the Bell-Astrom system is studied.After several improvements and debugging,the input variable order selection and genetic algorithm partial parameter adjustment methods are determined,its simulation verifies the feasibility of the modeling method in the furnace system.Then based on the actual unit test data,combined with the experience of Bell-Astrom system modeling methods,a fuzzy unit T-S model of the actual unit was established in MATLAB /Simulink,and the accuracy of the model was verified by calculating and comparing the mean square error of the simulation output data of the propeller speed,output power and main steam pressure under different modeling methods.(4)The basic principles of multivariate generalized predictive control algorithms are explained,and multi-model predictive control weighting strategies are formulated,combine T-S fuzzy model with predictive control algorithm and refer to Bell-Astrom system predictive controller design method,the multi-model generalized predictive controller for actual units is designed.Simulation tests of step disturbances and variable operating conditions were carried out to verify the effect of the predictive control algorithm in the marine furnace system.The results show that the improved T-S fuzzy modeling method proposed in this paper can effectively improve modeling efficiency and modeling accuracy.And the multivariable generalized predictive controller based on the model achieves faster and more stable control and better overall performance of steam power plants.