SOM Neural Network Fault Diagnosis Method of Polymerization Kettle Equipment Optimized by Improved PSO Algorithm

Table 8

Comparison between actual values and predicted values of SOM neural network optimized by improved PSO algorithm.

Type	Number of winning neurons
1~18	79	41	72	60	23	23	5	5	12	45	68	33	23	16	73	14	13	75
Fault class	1		1	3	2	4	3	3	4	1	3	2	4	2	1	2	2	4
19~36	23	18	60	73	24	19	65	1	1	59	73	14	13	23	34	5	12	45
Fault class	4	2	3	1	2	3	1	3	3	4	1	2	2	2		3	4	1
37~54	68	33	5	81	64	51	24	42	55	30	30	8	31	19	9	10	59	65
Fault class	3	2	3	1	1	3	2	4	1	4	4	2	4	3	2	3	4	1
55~72	58	80	1	5	13	75	45	43	41	72	60	23	57	5	12	45	2	23
Fault class	4	1	3	2	2	4	1			1	3	2	4	3	4	1	3	4
73~80	28	37	9	30	16	28	9	64
Fault class	3	3	2	4	2	3	2	1