Title :
Research on the fouling prediction of heat exchanger based on Relevance vector machine
Author :
Sun, Lingfang ; Saqi, Rina ; Xie, Honggang
Author_Institution :
Sch. of Autom. Eng., Northeast Dianli Univ., Jilin, China
Abstract :
The precision of fouling prediction has important significance to the design and countermeasure of heat exchanger. Whereas, the research on the fouling prediction based on Relevance Vector Machine (RVM) was carried on. The implement steps of prediction based on RVM were introduced and the predict curves were given. The simulation result showed that, the fouling prediction based on RVM could exactly predict the variation trend of fouling resistance. Compared to the support vector machine, RVM required dramatically fewer kernel functions and it could get higher prediction precision. So the prediction efficiency of RVM is better than that of SVM.
Keywords :
Bayes methods; heat exchangers; learning (artificial intelligence); maintenance engineering; mechanical engineering computing; prediction theory; support vector machines; fouling prediction; heat exchanger; kernel function; prediction curve; relevance vector machine; support vector machine; Automation; Bayesian methods; Heat engines; Kernel; Presses; Resistance; Support vector machines; Fouling Resistance; Prediction; Relevance Vector Machine;
Conference_Titel :
Intelligent Control and Automation (WCICA), 2010 8th World Congress on
Conference_Location :
Jinan
Print_ISBN :
978-1-4244-6712-9
DOI :
10.1109/WCICA.2010.5555170