Modeling And Nonlinear Control Of MGT-LiBr Chiller Distributed Energy System

Distributed Energy System is built on the concept of energy cascade utilization, it has many advantages such as being environmentally friendly and energy saving, efficiently and safely. Micro gas turbine based LiBr absorption chiller system is a typical combined cooling heating and power system (Abbreviation:MGT-LiBr CCHP), it is one of the mostly widely used small-scale distributed energy systems currently. Research on the static performance of the MGT-LiBr CCHP system has achieved many results. However, due to the complex structure and the multitudinous components of the MGT-LiBr CCHP system, there is still lack of knowledge on its dynamic mechanism and control rules. In order to obtain the static and dynamic performance of the MGT-LiBr CCHP system, rebuild its model quickly and design its control rules, Hammerstein model is delivered in this paper on the basic of description and analysis of the mechanism model of the MGT-LiBr CCHP system. Then a nonlinear predictive controller is designed for the cooling load following based on the Hammerstein model of the MGT-LiBr CCHP system and L1 adaptive controller is designed. Thus the study includes three parts:1.Description of the math model of the mainly components of the MGT-LiBr CCHP is given due to the ordinary differential equation. The thermal dynamic characteristics of the components are discussed and analyzed. Modelling of the mechanism model is introduced as well. The simulation experiment is carried out based on the mechanism model, the static and dynamic characteristics of the system under varying conditions is analyzed. Nonlinear measurement of the MGT-LiBr CCHP system is discuss in the paper.2. MGT-LiBr CCHP system’s structure is very complicated, thus the mechanism model of the MGT-LiBr CCHP system is hard to rebuild, considering of the lower calculate speed and the initial instable state, Hammerstein model is proposed for the MGT-LiBr CCHP system, which is also suitable for control rules design. The steady data are used to identify the nonlinear static parameters of the Hammerstein model. Stepwise regression is used to determine the order of the linear part before dynamic identification. Particle swarm optimization is used to get the parameters. Simulation experiment is carried out to analyze the characteristic of the models.3. A nonlinear predictive controller is designed for cooling load following based on the Hammerstein model. Two-step moving horizon optimization is used with Hammerstein model. The Generalized Predictive Control algorithm is firstly applied to the linear part, and then the required control variable is obtained by extracting the root of the nonlinear equation. Then its control effect and PID controller’s is compared. At last, L1 adaptive controller for the MGT-LiBr CCHP system is designed and simulation is carried out for validation.