A BP Neural Network Method for Grade Classification of Loose Damage in Semirigid Pavement Bases

Table 7

Part of the training sample set.

Attributes
Sample	Average absolute value amplitude	Root mean square amplitude	Amplitude standard deviation	Amplitude deviation	Waveform factor	Crest factor	Kurtosis factor	Pulse factor	Kurtosis
1	453.18	540.09	540.62	292271.33	1.40	8.33	3.17 E + 10	11.67	8.91
2	409.97	533.14	533.66	284795.67	1.40	8.29	3.08 E + 10	11.58	8.99
3	439.65	513.54	514.05	264244.13	1.40	8.49	2.97 E + 10	11.89	9.72
4	448.10	514.72	515.23	265457.96	1.39	8.27	2.77 E + 10	11.45	9.00
5	376.87	503.89	504.38	254402.69	1.38	8.28	2.58 E + 10	11.45	8.93
6	394.37	479.01	479.48	229899.25	1.40	8.49	2.41 E + 10	11.87	9.71
7	391.61	471.10	471.56	222370.98	1.40	8.53	2.27 E + 10	11.96	9.59
8	355.54	462.48	462.94	214309.79	1.41	8.65	2.2 E + 10	12.17	9.86
9	340.09	459.21	459.66	211290.84	1.41	8.76	2.18 E + 10	12.34	9.96
10	373.81	464.46	464.91	216141.95	1.42	8.82	2.23 E + 10	12.52	9.86