The Research of Fault Diagnosis of Nuclear Power Plant Based on ELM-AdaBoost.SAMME

Table 2

Simulation results of three algorithms.

Algorithm	Average accuracy of training samples	Average accuracy of testing samples	Average training time/s	Model parameters
ELM	0.8925	0.925	1.6018	Hidden layer number: 140
ELM-AdaBoost.SAMME	0.935	0.975	3.3401	Weak classifier number: 100; Hidden layer number: 140
BP	0.9638	0.918	46.7853	Hidden layer number: 45
BP-AdaBoost.SAMME	0.977	0.921	102.9676	Hidden layer number: 45; Hidden layer number: 100