Neural Networks and Fault Probability Evaluation for Diagnosis Issues

Table 5

Delayed diagnosis for fault 𝑓₃ with 𝑇=𝑡=1000 s.

Model of faults NNFM(𝑖)
Fault candidate	𝜌1𝑖(1000,1000,501)	𝜌1𝑖(1000,1000,501)	𝐷𝑖(1000,1000,501)	𝑃𝑖(1000,1000,501)
𝑓3	0.01	0.90	0.90	0.51
𝑓6	0.03	6.48	6.48	0.07
𝑓9	0.09	1.84	1.84	0.25
𝑓12	2.71	5.73	6.34	0.07
𝑓18	0.03	9.92	9.92	0.04
𝑓19	0.03	13.13	13.13	0.03