Research Article

Modeling a Thermochemical Reactor of a Solar Refrigerator by BaCl2-NH3 Sorption Using Artificial Neural Networks and Mathematical Symmetry Groups

Table 2

Weights and bias of the neural network.

Input layer
r1r2r3ThfVsVgtHDn

Neuron 13.26E + 001.63E + 000.00E + 004.81E − 021.40E − 042.26E − 04−9.14E − 011.39E − 015.96E − 02
Neuron 2−1.71E + 00−8.56E − 010.00E + 001.48E − 01−7.36E − 05−1.19E − 047.18E − 01−7.29E − 02−3.13E − 02
Neuron 34.29E − 012.15E − 010.00E + 005.96E − 021.85E − 052.97E − 055.72E − 011.83E − 027.85E − 03
Neuron 43.90E + 001.95E + 000.00E + 00−1.75E − 011.68E − 042.70E − 04−1.39E − 011.66E − 017.13E − 02
Neuron 5−4.39E − 01−2.19E−010.00E + 001.65E − 02−1.89E − 05−3.04E − 05−5.77E − 03−1.87E − 02−8.03E − 03
Neuron 63.60E + 001.80E + 000.00E + 00−3.76E − 011.55E − 042.49E − 041.59E + 001.53E − 016.58E − 02
Neuron 73.03E − 011.52E − 010.00E + 00−5.32E − 031.30E − 052.10E − 05−1.18E − 011.29E − 025.55E − 03
Neuron 8−6.36E − 02−3.18E − 020.00E + 00−8.44E − 02−2.74E − 06−4.41E − 061.37E − 01−2.71E − 03−1.16E − 03
Neuron 92.76E + 001.38E + 000.00E + 00−2.16E − 011.19E − 041.91E − 04−2.28E + 001.17E − 015.05E − 02

Hidden layer
Bias
123456789

b1−1.13E + 01−5.76E + 00−9.33E + 00−4.22E + 009.58E−01−5.05E + 00−1.03E + 004.56E + 004.26E + 00

Output layer
Inputs
123456789

Neuron 1−5.86E − 024.22E + 00−3.83E + 009.03E − 015.15E + 016.45E − 01−4.80E + 00−2.78E + 00−1.64E + 00
Neuron 2−4.09E − 013.27E + 00−3.77E + 008.74E − 015.41E + 018.69E − 01−1.03E + 01−7.84E − 012.38E − 01
Neuron 3−1.10E + 011.05E + 00−9.41E − 038.09E − 01−3.78E + 00−3.65E + 00−4.87E + 00−1.28E + 01−1.08E − 01
Neuron 43.88E + 00−2.67E + 013.95E + 01−3.04E + 01−5.10E + 019.09E−02−6.28E + 01−4.04E + 00−7.94E + 00
Neuron 58.20E − 024.34E − 01−2.80E − 011.70E + 005.02E + 003.54E − 01−2.67E + 00−2.41E + 002.14E − 01

Output layer
Bias
b26.19E + 016.08E + 014.83E + 013.76E + 016.93E + 00