Prediction of phenolic compounds and glucose content from dilute inorganic acid pretreatment of lignocellulosic biomass using artificial neural network modeling

Bioresources and Bioprocessing

Table 3 Weights and biases of the hidden and output layers used in the developed ANN model for the prediction of C_Glc and C_Phe

Node, j	Weights and biases of the hidden layer							Weights and biases of the output layer
Node, j	IW_j,1	IW_j,2	IW_j,3	IW_j,4	IW_j,5	IW_j,6	b_1,j	LW_1,j	LW_2,j	b_2,k
1	0.4211	0.5806	0.1236	0.2909	0.1691	0.5445	− 0.0095	0.0448	0.4378	− 0.0381
2	0.3931	0.1473	0.2595	0.4280	0.2980	0.4272	− 0.0393	0.3972	0.4253	− 0.0051
3	0.3056	0.1740	0.3713	0.1522	0.1229	0.1367	− 0.0287	0.0356	0.2299
4	0.5184	0.0507	0.2491	0.0508	0.3027	0.4435	− 0.0379	0.6842	0.1607
5	0.3112	0.4807	0.7085	0.2731	0.3305	0.6146	0.0394	0.2159	0.0281
6	0.1854	0.5232	0.0776	0.4782	0.3439	0.1783	− 0.0306	0.1249	0.4419
7	0.0370	0.2334	0.0454	0.3650	0.2354	0.4857	0.0372	0.3608	0.0639
8	0.2985	0.2512	0.3173	0.1907	0.1834	0.5006	0.0361	0.5057	0.4299
9	0.3070	0.4097	0.0005	0.5172	0.3769	0.2365	− 0.0184	0.0140	0.3469
10	0.3136	0.2505	0.1438	0.0805	0.0552	0.1178	− 0.0320	0.2159	0.6152
11	0.6273	0.0770	0.3091	0.3422	0.3044	0.3722	0.0394	0.6076	0.3147
12	0.5133	0.5409	0.0001	0.4964	0.2515	0.1423	0.0371	0.5611	0.2426

j is the neuron node in the hidden layer (from j = 1 to 12). IW_j,i (i = 1, 2, 3, 4, 5, and 6) is the neural net weight to jth neuron of the hidden layer from ith input variable; and LW_k,j (k = 1 and 2) is the neural net weight to kth output variable from jth neuron of the hidden layer. b_1,j is the bias of input variables, and b_2,k is the bias of output layer