An Artificial Neural Networks Model for Predicting Permeability Properties of Nano Silica–Rice Husk Ash Ternary Blended Concrete

International Journal of Concrete Structures and Materials

Table 5 Summary of six NNMs for permeability properties of concrete.

Name	No. of PEs in HL^b	Training set		Validation set		Testing set		All	Epochs
Name	No. of PEs in HL^b	MSE	R	MSE	R	MSE	R	R	Epochs
NNM1	1	3.5103	0.95412	2.53146	0.95263	2.47188	0.97263	0.95979	224
NNM1	1	Output = 0.94 × target + 0.38		Output = 0.87 × target + 0.43		Output = 0.97 × target + 0.31		Output = 0.93 × target + 0.37	224
NNM2	6	1.60119	0.98889	1.46301	0.98451	1.38133	0.97156	0.98165	16
NNM2	6	Output = 0.99 × target + 0.052		Output = 0.94 × target + 0.18		Output = 1 × target +0.071		Output = 0.99 × target + 0.05	16
NNM3	11	1.08359	0.98449	1.15492	0.97645	1.25615	0.98985	0.98359	16
NNM3	11	Output = 0.98 × target + 0.096		Output = 0.99 × target + 0.21		Output = 0.98 × target + 0.11		Output = 0.98 × target + 0.12	16
NNM4	16	0.80060	0.98219	0.96029	0.99049	0.66266	0.97940	0.98402	15
NNM4	16	Output = 0.99 × target + 0.07		Output = 0.99 × target + 0.185		Output = 1 × target +0.08		Output = 1 × target + 0.049	15
NNM5	20	0.47303	0.9998	0.50148	0.99040	0.40688	0.99448	0.99303	14
NNM5	20	Output = 1 × target + 0.001		Output = 1 × target + 0.32		Output = 0.94 × target + 0.32		Output = 1 × target + 0.04	14
NNM6	24	0.97703	0.97889	1.00621	0.96456	0.70978	0.97889	0.97411	10
NNM6	24	Output = 97 × target +0.21		Output = 0.97 × target + 0.21		Output = 0.98 × target + 0.11		Output = 0.98 × target + 0.21	10