Conjugate Cellular Automata and Neural Network Approach: Failure Load Prediction of Masonry Panels

Table 4

Comparison between NN, FEA, and YL prediction with the experimental data (Expt.).

Panel	Expt. (kN/m^2)	YL (kN/m^2)	FEA (kN/m^2)	NN (kN/m^2)	Panel	Expt. (kN/m^2)	YL (kN/m^2)	FEA (kN/m^2)	NN (kN/m^2)
1120	10.9	13.16	9.65	10.975	1150	10.7	11.05	8.83	10.727
1135	6.5	9.87	6.85	3.710	1153	6.7	9.06	7.04	5.973
1116	7.9	8.88	6.67	6.551	1237	6.5	8.24	6.2	6.406
1126	6.4	8.88	6.67	6.551	1148	15.8	22.09	15.37	16.057
1190	6.4	8.46	6.65	6.480	1211	12.8	20.25	14.9	12.900
1108	5.2	6.66	4.87	5.356	1246	11.8	19.29	15.09	12.103
1109	3.2	3.8	2.77	3.071	1170	8.6	9.96	7.53	8.938
1187	2.8	3.38	2.33	5.303	1178	5.2	8.76	5.93	4.087
1094	3.6	5.04	4.08	4.246	1224	4.8	8.11	5.49	7.872
1110	5	5.04	4.08	4.246	1149	6.6	7.7	5.63	5.864
1095	2.9	2.77	1.95	3.317	1162	4.3	6.37	4.5	4.504
1171	2.2	2.41	1.67	3.593	1231	3.9	5.86	3.92	6.272
1121	5.8	6.88	4.69	5.424	ART01	4.4	—	4.56	4.290
1123	4.4	5.35	3.04	1.887	SB01	2.8	3.164	2.464	2.968
1117	4.3	4.98	3.02	4.030	SB05	2.7	3.159	2.457	3.184
1203	3.4	4.74	2.97	2.580	Panels with an opening
1107	5	3.43	2.31	5.530	SB06	7.5	8.7	6.75	5.920
1111	4.3	2.06	1.17	2.528	Panel 1	1.5	—	1.31	1.769
1201	2.1	1.87	1.01	3.044	Panel 2	1.2	—	1.22	1.503
1096	2.8	1.65	1.89	2.227	Panel 3	2	—	1.24	2.125
1097	1.8	1.5	0.91	3.067	ART02	2.75	—	2.39	2.487
1157	1.4	1.35	0.73	1.840	ART03	2.6	—	2.6	2.933
1172	20.9	31.35	23.75	20.372	ART04	2.5	—	2.21	2.557
1192	18.8	28.08	23.6	18.862	ART06	2.5	—	2.69	2.437
1261	15	27.26	23.59	15.238	SB02	2.4	2.592	1.824	2.461
1169	9.5	14.4	11.66	10.059	SB03	2.3	2.438	1.587	2.478
1173	8	12.31	9.49	5.274	SB04	2.2	2.574	1.804	2.602
1244	8	11.28	8.63	8.112	SB07	5.5	6.985	5.17	5.514