Model-based Optimization Of Vapor Compression Cycles

Vapor compression refrigeration cycle is a nonlinear process with strong interactions among each working device, constantly changing work conditions and strong inertia. Due to the urgency on energy saving and environmental protection, refrigeration cycle performance simulation, modeling, control and optimization had draw a lot of research interests. In this thesis, a model-based global optimization method of vapor compression cycles is presented from the system engineering point of view. First, through the characteristic analysis, a hybrid modeling method is proposed, and hybrid models for two-phase flow heat exchangers (condensers and evaporators) as well as compressor are developed. The main procedures for hybrid modeling includes: 1) Based on the energy and material balance, and thermodynamic principles to formulate the process fundamental governing equations; 2) Select input/output (I/O) variables responsible to the system performance which can be measured and controlled; 3) Represent those variables existing in the original equations but are not measurable as simple functions of selected I/Os or constants; 4) Obtaining a single equation which can correlate system inputs and outputs; and 5) Identify unknown parameters by linear or nonlinear least squares methods. This modeling approach has three inherent advantages: 1) they incorporate theoretical knowledge; 2) they can be extrapolated over a wider range of operating conditions than empirical models; and 3) they require less computation effort than theoretical models. The results of simulations and experiments show that each of the component models is not only simple, but also can accurately predict their performance. These models are very suitable for the real time control and optimization of industrial processes. Secondly, through the analysis on the variables which affect the system performance, a mixed-integer nonlinear constraint optimization problem of the overall system energy is formulated based on mathematical models. At last, a modified genetic algorithm is used to solve the optimization problems and a detailed calculation procedure and the comparison results with the traditional method are given. The simulation results show that the proposed method indeed improves the system performance significantly. The main contribution of this thesis is to propose a hybrid modeling method suitable for two-phase heat exchangers and a general systematic approach in optimizing the energy consumption of overall vapor compression systems instead of optimizing individual components or parts of systems. The proposed method can be further extended to the application on the performance optimization of other types of refrigeration cycles and industry processes.