Artificial Hydrocarbon Networks Fuzzy Inference System

Algorithm 1

Training algorithm for artificial hydrocarbon networks.

(1) Initialize AHN=∅

(2) For each C_𝑖  do

(3) Initialize 𝑟^𝑗𝑘 randomly under the input domain

(4) While stop condition is not reached do

(5)  Split (𝑥,𝑦)-pairs into 𝑝 clusters 𝑦_𝑗 using 𝑟^𝑗𝑘

(6)  For each cluster 𝑦_𝑗  do

(7)   Create a CH-molecule using criterion (6)

(8)   Obtain hydrogen values of molecule M_𝑗 using LSE method

(9)   Calculate least square error 𝐸_𝑗 between 𝑦_𝑗 and M_𝑗

(10)  End for

(11)  Update intermolecular distances 𝑟^𝑗𝑘 using (5)

(12) End while

(13) Update AHN←AHN∪C_𝑖

(14) Update (𝑥,𝑦)-pairs with (𝑥,𝑦−AHN)-pairs

(15) End for

(16) Obtain stoichiometric coefficients of C_𝑖 compounds using LSE method

(17) Return AHN