Accès libre

Advanced Ai Tools for Predicting Mechanical Properties of Self-Compacting Concrete

 et   
10 janv. 2025
À propos de cet article

Citez
Télécharger la couverture

Figure 1.

ANFIS network designed for the SCC strength modeling [28]
ANFIS network designed for the SCC strength modeling [28]

Figure 2.

Hyperplanes of SVR [30]
Hyperplanes of SVR [30]

Figure 3.

ANN architecture [23]
ANN architecture [23]

Figure 4.

Relative importance of each parameter
Relative importance of each parameter

Figure 5.

Training phase results of ANFIS for CS28
Training phase results of ANFIS for CS28

Figure 6.

Testing phase results of ANFIS for CS28
Testing phase results of ANFIS for CS28

Figure 7.

Training phase results of ANFIS for TS28
Training phase results of ANFIS for TS28

Figure 8.

Testing phase results of ANFIS for TS28
Testing phase results of ANFIS for TS28

Figure 9.

CS28 results from prediction models
CS28 results from prediction models

Figure 10.

TS28 results from prediction models
TS28 results from prediction models

SVM parameters used in the present study [20]

SVM Parameter Adoption/Values
Kernel function RBF
Scaling factor 1
Method Quadratic programming
Support Vectors 13x5
Bias -0.012

Correlation matrix for TS28

Variables SP PC PPF SF SFD SFT VF TS28
SP 1.00 0.14 0.00 0.00 0.42 −0.36 −0.32 0.71
PC 0.14 1.00 0.02 0.02 0.36 −0.36 −0.44 0.69
PPF 0.00 0.02 1.00 −0.01 −0.67 0.70 0.66 0.21
SF 0.00 0.02 −0.01 1.00 −0.30 0.23 0.31 0.29
SFD 0.42 0.36 −0.67 −0.30 1.00 −0.98 −0.96 0.26
SFT −0.36 −0.36 0.70 0.23 −0.98 1.00 0.97 −0.23
VF −0.32 −0.44 0.66 0.31 −0.96 0.97 1.00 −0.25
TS28 0.71 0.69 0.21 0.29 0.26 −0.23 −0.25 1.00

Performance Matrix for CS28

Models Index of Agreement (IOA) Akaike Information Criterion (AIC) Skill Score (SS) Symmetric Uncertainty (SU)
ANFIS 0.50 256.34 0.50 0.01
ANN 0.64 232.34 0.64 0.29
SVM 0.96 68.33 0.96 0.93
MLR 0.51 256.04 0.51 0.01
GEP 0.62 159.31 0.58 0.73

Performance Matrix for TS28

Models Index of Agreement (IOA) Akaike Information Criterion (AIC) Skill Score (SS) Symmetric Uncertainty (SU)
ANFIS 0.67 −59.96 0.67 0.33
ANN −0.54 50.01 −0.54 −2.07
SVM 0.81 −99.25 0.81 0.61
MLR 0.87 −127.64 0.87 0.74
GEP 0.62 −51.32 0.62 0.25

Correlation matrix for CS28

VARIABLES SP PC PPF SF SFD SFT VF CS28
SP 1.00 0.14 0.00 0.00 0.42 −0.36 −0.32 0.68
PC 0.14 1.00 0.02 0.02 0.36 −0.36 −0.44 0.72
PPF 0.00 0.02 1.00 −0.01 −0.67 0.70 0.66 0.16
SF 0.00 0.02 −0.01 1.00 −0.30 0.23 0.31 0.32
SFD 0.42 0.36 −0.67 −0.30 1.00 −0.98 −0.96 0.28
SFT −0.36 −0.36 0.70 0.23 −0.98 1.00 0.97 −0.25
VF −0.32 −0.44 0.66 0.31 −0.96 0.97 1.00 −0.26
CS28 0.68 0.72 0.16 0.32 0.28 −0.25 −0.26 1.00

Initial values of ANFIS GA parameters [28]

Parameters Values
Population size 20
Iterations 1000
Crossover rate 0.70
Mutation rate 0.50
Inversion rate 0.10
Selection pressure 8.0
Gamma 0.20

Dataset for training and testing of different models

SP (kg) PC (kg) PPF (kg) SF (kg) SFD (Dia) SFT (Sec) VF (Sec) CS28 (MPa) TS28 (MPa) Water (kg) FA (kg) CA1 20-10 mm (kg) CA2 10-4.75 mm (kg)
9 600 0 0 780 2 7 36.11 4.41 240 810 365 365
9 600 1.5 3 665 5 13 39.95 4.79 240 810 365 365
9 600 3 6 630 6 14 41.1 5.26 240 810 365 365
9.225 615 0 0 790 1.5 6 39.55 4.67 246 810 365 365
9.225 615 1.535 3.07 730 3 9 42.33 5.66 246 810 365 365
9.225 615 3.07 6.15 690 4 11 46.31 6.12 246 810 365 365
9.45 630 0 3.15 750 2.5 8 45.52 6.03 252 810 365 365
9.45 630 1.575 6.3 730 3 8 49.16 6.88 252 810 365 365
9.45 630 3.15 0 725 3 9 46.46 6.28 252 810 365 365
12 600 0 6 760 2 8 44.12 5.86 240 810 365 365
12 600 1.5 0 765 2 7 41.68 5.53 240 810 365 365
12 600 3 3 710 4 12 45.09 6.22 240 810 365 365
12.3 615 0 3.07 775 2 7 48.03 6.74 246 810 365 365
12.3 615 1.535 6.15 725 3 9 53.11 7.26 246 810 365 365
12.3 615 3.07 0 710 4 10 47.36 6.67 246 810 365 365
12.6 630 0 6.3 810 1 6 57.16 7.8 252 810 365 365
12.6 630 1.575 0 785 2 6 54.3 7.52 252 810 365 365
12.6 630 3.15 3.15 750 3 9 56.44 7.76 252 810 365 365
3.3 565 0 0 690 2 11 52.8 3.9 186 750 58.5 391.5
5 566.7 0 0 680 2.2 16 57.3 4 170 760 58.5 391.5
5.1 566.7 0 39 665 2.8 18 56.9 5.9 170 760 58.5 391.5
5.3 566.7 4.55 39 670 3.1 19 61.7 6.9 170 760 58.5 391.5
5.7 566.7 6.825 39 660 3.3 18 58.8 7.2 170 760 58.5 391.5
6.2 566.7 9.1 39 645 4.2 20 56.7 6.9 170 760 58.5 391.5
10 500 0 0 700 2.4 7 37.21 4.25 190 1005 310 310
10 500 0 157 600 4 12 51.31 7.2 190 1005 310 310
10 500 0 117.75 620 3.8 11 50.24 6.25 190 1005 310 310
10 500 0 78.5 640 3.5 10.5 50.45 6.1 190 1005 310 310
10 500 0 39.25 650 3.5 10.5 45.64 4.8 190 1005 310 310
10 500 0 0 670 3 8.3 34.87 4.1 190 1005 310 310
10 500 0.455 157 580 5 12.7 40.52 5.8 190 1005 310 310
10 500 0.455 117.75 600 4 12 45.78 5.8 190 1005 310 310
10 500 0.455 78.5 620 3.5 11 50.44 6.25 190 1005 310 310
10 500 0.455 39.25 630 3.5 10.6 47.86 5.85 190 1005 310 310
10 500 0.455 0 640 3.2 10.5 47.84 5.75 190 1005 310 310
10 500 0.91 157 560 5.5 13 64.1 7.25 190 1005 310 310
10 500 0.91 117.75 570 5.5 12.5 54.98 6.15 190 1005 310 310
10 500 0.91 78.5 590 5 11.5 47.64 5.7 190 1005 310 310
10 500 0.91 39.25 610 4.5 11 46.09 5.75 190 1005 310 310
10 500 0.91 0 620 4 11 43.96 3.8 190 1005 310 310
10 500 0 0 720 2.2 6.8 39.04 4.3 190 1005 310 310
10 500 0 157 620 4 11.6 66.69 7.95 190 1005 310 310
10 500 0 117.75 630 4 11 53.91 6.5 190 1005 310 310
10 500 0 78.5 640 3.5 10.5 52.31 6.5 190 1005 310 310
10 500 0 39.25 660 3.5 10.5 47.96 5.7 190 1005 310 310
10 500 0 0 680 2.5 8.1 39.53 4.6 190 1005 310 310
10 500 0.455 157 600 4.5 12.5 52.27 6.45 190 1005 310 310
10 500 0.455 117.75 620 4 12 62.67 7.35 190 1005 310 310
10 500 0.455 78.5 630 4 11 56.92 6.5 190 1005 310 310
10 500 0.455 39.25 640 3.5 10.4 55.13 6.45 190 1005 310 310
10 500 0.455 0 650 3.5 10.2 53.21 6.2 190 1005 310 310
10 500 0.91 157 580 5.5 12.5 69.12 8.1 190 1005 310 310
10 500 0.91 117.75 590 5.5 12 55.68 6.8 190 1005 310 310
10 500 0.91 78.5 600 5 11.8 50.05 6.4 190 1005 310 310
10 500 0.91 39.25 620 4 11 46.5 5.9 190 1005 310 310
10 500 0.91 0 630 4 10.6 46.05 4.4 190 1005 310 310
10 500 0 0 740 2 6.8 42.83 4.95 190 1005 310 310
10 500 0 157 630 4 11.6 68.14 8.05 190 1005 310 310
10 500 0 117.75 630 4 11 64.79 7.5 190 1005 310 310
10 500 0 78.5 650 3.2 10.5 56.55 6.65 190 1005 310 310
10 500 0 39.25 660 3 10.5 49.82 5.9 190 1005 310 310
10 500 0 0 700 2.5 8.1 46.72 4.75 190 1005 310 310
10 500 0.455 157 620 4.5 12.5 60.22 7 190 1005 310 310
10 500 0.455 117.75 630 4 12 69.71 8.15 190 1005 310 310
10 500 0.455 78.5 640 4 11 64.52 7.4 190 1005 310 310
10 500 0.455 39.25 650 3.5 10.4 56.7 6.6 190 1005 310 310
10 500 0.455 0 670 3 10.2 53.73 6.55 190 1005 310 310
10 500 0.91 157 600 5 12.5 69.79 8.15 190 1005 310 310
10 500 0.91 117.75 610 5 12 56.67 7.75 190 1005 310 310
10 500 0.91 78.5 620 4.5 11.8 52.48 6.8 190 1005 310 310
10 500 0.91 39.25 640 3.5 11 49.6 6.15 190 1005 310 310
10 500 0.91 0 650 3.5 10.6 49.4 5 190 1005 310 310

GEP parameters [27]

Parameters Values
Population size 30
Genes per chromosome 3
Gene head length 9
Functions +, -./, *,^
Gene tail length 12
Mutation rate 0.05
Inversion rate 0.1
Gene transposition rate 0.1
One-point recombination rate 0.3
Two-point recombination rate 0.3
Gene recombination rate 0.1
Fitness function R≥0.7
Langue:
Anglais
Périodicité:
4 fois par an
Sujets de la revue:
Architecture et design, Architecture, Architectes, bâtiments