Font Size: a A A

Evaporator In The Hvac System Modeling And Simulation

Posted on:2007-10-18Degree:MasterType:Thesis
Country:ChinaCandidate:H L YuFull Text:PDF
GTID:2208360185483158Subject:Detection Technology and Automation
Abstract/Summary:PDF Full Text Request
Heating, Ventilation, and Air-Conditioning (HVAC) systems are complex and non-linear. With the increasingly lacking of energy sources, the study about their math model and optimized control was gradually concerned by researchers. Especially for china, which was highly short of energy source, it was truly necessary to lucubrate the modeling and control of the systems that consumed large energy sources.The conventional modeling method in HVAC systems was based on mathematics mechanism, which was complex and difficultly applied in project. In this paper, based on the neural network identify and ameliorated BP arithmetic, an evaporator model was simulated which was aimed at the practicality apply.1) The paper introduced the working process of HVAC systems and the classical math model and simulation about evaporator, such as finite difference model, lumped parameters model and distributing parameter model. The work of researchers was listed, including abroad and at home.2) Discussed the working principle of thermal and environmental variables such as temperature, pressure, enthalpy, etc. Firstly, the input and output parameters of the evaporator were radicated, and the distributing parameter evaporator model was established by classical math mechanism method. Then the emluator interface was compiled, which can be conveniently inputted the parameters. Further more, the data between experiment and simulation was compared and analyzed.3) Neural Network was introduced, which was successfully applied in the field of identification, intelligent simulation predicts and so on. Because of Neural Network's unique features in dealing and adapting to non-linear information, it was particular and...
Keywords/Search Tags:lumped parameter model, enthalpy, intelligent simulation, identification, Neural Network
PDF Full Text Request
Related items