Design of Jetty Piles Using Artificial Neural Networks

Table 6

Example of training samples.

(a)

Cases	Input neuron (load and information of each pile)
	1; (Load)	2; (Pile 1)	3; (Pile 2)	4; (Pile 3)	5; (Pile 4)	⋯	11; (Pile 10)	12; (Pile 11)	13; (Pile 12)
1	0.7	1	1	1	1	⋯	1	1	1
2	0.7	2	2	2	2	⋯	2	2	2
3	0.7	1	1	1	1	⋯	1	1	1
⋮						⋮
48	1.1	1	1	1	2	⋯	1	1	1
49	1.1	1	1	2	0	⋯	1	1	1
50	1.1	1	1	1	2	⋯	1	1	1

(b)

Cases	Output neuron (stress ratio of each pile)
	1; (Pile 1)	2; (Pile 2)	3; (Pile 3)	4; (Pile 4)	5; (Pile 5)	⋯	11; (Pile 11)	12; (Pile 12)
1	−0.364	−0.363	−0.364	0.302	−0.727	⋯	−0.307	−0.305
2	−0.877	−0.877	−0.877	−0.847	−0.847	⋯	−0.771	−0.771
3	−0.547	−0.548	−0.547	0.404	0.404	⋯	−0.399	−0.397
48	−0.750	−0.751	−0.750	0.737	0.737	⋯	0.573	0.574
49	−0.788	−0.787	−0.788	0.856	0.856	⋯	0.651	0.651
50	−0.863	−0.864	−0.863	0.831	0.830	⋯	0.613	0.616