| Summary: | 碩士 === 國立臺北科技大學 === 能源與冷凍空調工程系碩士班 === 103 === ABSTRACT
Title：Applying Support Vector Regression to Predict Chiller Performance of
Air - Conditioning System
Pages：72
School：National Taipei University of Technology
Department：Energy and Refrigerating Air-Conditioning Engineering
Time：June, 2015 Degree：Master
Researcher：Si-Xian-Lu Advisor：Yung-Chung Chang
Keywords：coefficient of performance (C.O.P) simulation, regression analysis, backpropagation network, Support Vector Regression
Three methods are applied in this study to predict chiller performance.They are linear regression, neural network and Support Vector Regression. The C.O.P models of water cooled chiller and air cooled chiller are established by using these three methods. After that, the simulated results are compared and the performance of C.O.P models is improved by using these three methods under the same baseline.
To compare two different cases, data of air cooled chiller spread completely and data of water cooled chiller spread scatter. Two different cases are simulated by these three methods. We compare accurate and error with three methods .Case1 is the C.O.P model of the air chiller, .The results show that prediction and simulation are accurate by these three methods .Case2 is the C.O.P model of the water cooled chiller, The results show that Support Vector Regression method is the most accurate method among the three methods.
|
|---|
