Author Contributions
Conceptualization, K.M. and N.I.C.; methodology, K.M.; software, K.M.; validation, N.I.C., M.A.Z.R. and Z.A.K.; formal analysis, Z.A.K.; writing—original draft preparation, K.M.; writing—review and editing, N.I.C., Z.A.K. and M.A.Z.R.; project administration, K.M.C. and A.H.M.; funding acquisition, K.M.C. and A.H.M. All authors have read and agreed to the published version of the manuscript.
Figure 1.
Graphical abstract of the proposed study.
Figure 1.
Graphical abstract of the proposed study.
Figure 3.
Fitness plots for AO with respect to population sizes. (a) T = 1000. (b) T = 1500. (c) T = 2000.
Figure 3.
Fitness plots for AO with respect to population sizes. (a) T = 1000. (b) T = 1500. (c) T = 2000.
Figure 4.
Fitness plots for AO with respect to generations. (a) Np = 30. (b) Np = 40. (c) Np = 50.
Figure 4.
Fitness plots for AO with respect to generations. (a) Np = 30. (b) Np = 40. (c) Np = 50.
Figure 5.
Fitness plots for AO w.r.t population sizes. (a) T = 1000. (b) T = 1500. (c) T = 2000. (d) T = 1000. (e) T = 1500. (f) T = 2000. (g) T = 1000. (h) T = 1500. (i) T = 2000.
Figure 5.
Fitness plots for AO w.r.t population sizes. (a) T = 1000. (b) T = 1500. (c) T = 2000. (d) T = 1000. (e) T = 1500. (f) T = 2000. (g) T = 1000. (h) T = 1500. (i) T = 2000.
Figure 6.
Fitness plots for AO with respect to population sizes. (a) Np = 30. (b) Np = 30. (c) Np = 30. (d) Np = 40. (e) Np = 40. (f) Np = 40. (g) Np = 50. (h) Np = 50. (i) Np = 50.
Figure 6.
Fitness plots for AO with respect to population sizes. (a) Np = 30. (b) Np = 30. (c) Np = 30. (d) Np = 40. (e) Np = 40. (f) Np = 40. (g) Np = 50. (h) Np = 50. (i) Np = 50.
Figure 7.
Statistical analyses plots of AO, AOA, SCA, and RSA for Np = 50, T = 2000. (a) Noise level = 0.04. (b) Noise level = 0.06. (c) Noise level = 0.08.
Figure 7.
Statistical analyses plots of AO, AOA, SCA, and RSA for Np = 50, T = 2000. (a) Noise level = 0.04. (b) Noise level = 0.06. (c) Noise level = 0.08.
Figure 8.
Mann-Whitney U test between AO and AOA where * p < 0.01.
Figure 8.
Mann-Whitney U test between AO and AOA where * p < 0.01.
Figure 9.
Mann-Whitney U test between AO and SCA where * p < 0.01.
Figure 9.
Mann-Whitney U test between AO and SCA where * p < 0.01.
Figure 10.
Mann-Whitney U test between AO and RSA where * p < 0.01.
Figure 10.
Mann-Whitney U test between AO and RSA where * p < 0.01.
Table 1.
Parameter-tuning cases for AO analysis.
Table 1.
Parameter-tuning cases for AO analysis.
Case No. | β Value | μ Value |
---|
1 | 0.9 | 0.9 |
2 | 0.9 | 0.5 |
3 | 0.9 | 0.1 |
4 | 0.5 | 0.9 |
5 | 0.5 | 0.5 |
6 | 0.5 | 0.1 |
7 | 0.1 | 0.9 |
8 | 0.1 | 0.5 |
9 | 0.1 | 0.1 |
Table 2.
AO parameter analysis for case 1.
Table 2.
AO parameter analysis for case 1.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 3.
AO parameter analysis for case 2.
Table 3.
AO parameter analysis for case 2.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 4.
AO parameter analysis for case 3.
Table 4.
AO parameter analysis for case 3.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 5.
AO parameter analysis for case 4.
Table 5.
AO parameter analysis for case 4.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 6.
AO parameter analysis for case 5.
Table 6.
AO parameter analysis for case 5.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 7.
AO parameter analysis for case 6.
Table 7.
AO parameter analysis for case 6.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 8.
AO parameter analysis for case 7.
Table 8.
AO parameter analysis for case 7.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 9.
AO parameter analysis for case 8.
Table 9.
AO parameter analysis for case 8.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 10.
AO parameter analysis for case 9.
Table 10.
AO parameter analysis for case 9.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 11.
AO parameter tuning mean value analysis.
Table 11.
AO parameter tuning mean value analysis.
Cases | Mean Fitness Value |
---|
Case 1 | |
Case 2 | |
Case 3 | |
Case 4 | |
Case 5 | |
Case 6 | |
Case 7 | |
Case 8 | |
Case 9 | |
Table 12.
AO analysis with respect to generation and population sizes at 0.04 noise variance.
Table 12.
AO analysis with respect to generation and population sizes at 0.04 noise variance.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 13.
AO analysis with respect to generation and population sizes at 0.06 noise variance.
Table 13.
AO analysis with respect to generation and population sizes at 0.06 noise variance.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 14.
AO analysis with respect to generation and population sizes at 0.08 noise variance.
Table 14.
AO analysis with respect to generation and population sizes at 0.08 noise variance.
Generations (T) | Population (Np) | Average Fitness | Best Fitness | Worst Fitness | STD |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 15.
Comparison of AO with AOA, SCA, and RSA against the true values for the CAR model at 0.04 noise variance.
Table 15.
Comparison of AO with AOA, SCA, and RSA against the true values for the CAR model at 0.04 noise variance.
Algorithm | Generations (T) | Population (Np) | Design Parameters | Best Fit |
---|
r1 | r2 | s1 | s2 |
---|
AO | 1000 | 30 | 1.362 | 0.767 | 1.692 | 2.338 | |
40 | 1.357 | 0.765 | 1.706 | 2.308 | |
50 | 1.353 | 0.764 | 1.712 | 2.298 | |
1500 | 30 | 1.361 | 0.759 | 1.702 | 2.319 | |
40 | 1.356 | 0.768 | 1.704 | 2.316 | |
50 | 1.357 | 0.759 | 1.703 | 2.302 | |
2000 | 30 | 1.366 | 0.771 | 1.723 | 2.321 | |
40 | 1.359 | 0.760 | 1.686 | 2.329 | |
50 | 1.348 | 0.757 | 1.717 | 2.284 | |
RSA | 1000 | 30 | 1.190 | 0.599 | 1.740 | 1.855 | |
40 | 1.376 | 0.769 | 1.619 | 2.418 | |
50 | 1.201 | 0.569 | 1.371 | 2.042 | |
1500 | 30 | 1.281 | 0.734 | 1.663 | 2.178 | |
40 | 1.294 | 0.649 | 1.384 | 2.381 | |
50 | 1.127 | 0.579 | 1.704 | 1.767 | |
2000 | 30 | 1.272 | 0.704 | 1.174 | 2.648 | |
40 | 1.381 | 0.769 | 1.526 | 2.510 | |
50 | 1.227 | 0.708 | 1.719 | 2.051 | |
SCA | 1000 | 30 | 1.341 | 0.715 | 1.646 | 2.278 | |
40 | 1.354 | 0.774 | 1.725 | 2.332 | |
50 | 1.394 | 0.766 | 1.678 | 2.397 | |
1500 | 30 | 1.376 | 0.759 | 1.662 | 2.443 | |
40 | 1.364 | 0.779 | 1.572 | 2.497 | |
50 | 1.347 | 0.742 | 1.626 | 2.355 | |
2000 | 30 | 1.350 | 0.753 | 1.694 | 2.334 | |
40 | 1.326 | 0.782 | 1.600 | 2.389 | |
50 | 1.312 | 0.704 | 1.691 | 2.205 | |
AOA | 1000 | 30 | 1.346 | 0.866 | 1.494 | 2.613 | |
40 | 1.449 | 0.881 | 1.465 | 2.833 | |
50 | 1.353 | 0.769 | 1.555 | 2.519 | |
1500 | 30 | 1.394 | 0.762 | 1.791 | 2.250 | |
40 | 1.345 | 0.726 | 1.649 | 2.358 | |
50 | 1.385 | 0.800 | 1.613 | 2.512 | |
2000 | 30 | 1.331 | 0.677 | 1.455 | 2.409 | |
40 | 1.383 | 0.724 | 1.569 | 2.433 | |
50 | 1.346 | 0.742 | 1.763 | 2.166 | |
True Values | 1.350 | 0.750 | 1.680 | 2.320 | 0 |
Table 16.
Comparison of AO with AOA, SCA, and RSA against the true values for the CAR model at 0.06 noise variance.
Table 16.
Comparison of AO with AOA, SCA, and RSA against the true values for the CAR model at 0.06 noise variance.
Algorithm | Generations (T) | Population (Np) | Design Parameters | Best Fit |
---|
r1 | r2 | s1 | s2 |
---|
AO | 1000 | 30 | 1.371 | 0.771 | 1.701 | 2.339 | |
40 | 1.369 | 0.777 | 1.715 | 2.336 | |
50 | 1.368 | 0.773 | 1.700 | 2.340 | |
1500 | 30 | 1.366 | 0.771 | 1.744 | 2.300 | |
40 | 1.362 | 0.761 | 1.733 | 2.291 | |
50 | 1.356 | 0.769 | 1.718 | 2.305 | |
2000 | 30 | 1.367 | 0.770 | 1.711 | 2.327 | |
40 | 1.364 | 0.776 | 1.723 | 2.317 | |
50 | 1.363 | 0.772 | 1.706 | 2.333 | |
RSA | 1000 | 30 | 1.200 | 0.657 | 1.159 | 2.531 | |
40 | 1.202 | 0.600 | 1.789 | 1.882 | |
50 | 1.183 | 0.580 | 1.425 | 2.118 | |
1500 | 30 | 1.120 | 0.582 | 1.584 | 1.850 | |
40 | 1.253 | 0.645 | 1.091 | 2.681 | |
50 | 1.265 | 0.683 | 1.431 | 2.396 | |
2000 | 30 | 1.102 | 0.600 | 1.888 | 1.557 | |
40 | 1.353 | 0.762 | 1.341 | 2.726 | |
50 | 1.342 | 0.758 | 1.349 | 2.610 | |
SCA | 1000 | 30 | 1.341 | 0.715 | 1.646 | 2.278 | |
40 | 1.354 | 0.774 | 1.725 | 2.332 | |
50 | 1.394 | 0.766 | 1.678 | 2.397 | |
1500 | 30 | 1.376 | 0.759 | 1.662 | 2.443 | |
40 | 1.364 | 0.779 | 1.572 | 2.497 | |
50 | 1.347 | 0.742 | 1.626 | 2.355 | |
2000 | 30 | 1.350 | 0.753 | 1.694 | 2.334 | |
40 | 1.326 | 0.782 | 1.600 | 2.389 | |
50 | 1.312 | 0.704 | 1.691 | 2.205 | |
AOA | 1000 | 30 | 1.427 | 0.870 | 1.470 | 2.808 | |
40 | 1.430 | 0.811 | 1.295 | 2.854 | |
50 | 1.315 | 0.737 | 1.738 | 2.167 | |
1500 | 30 | 1.419 | 0.789 | 1.446 | 2.662 | |
40 | 1.339 | 0.779 | 1.702 | 2.328 | |
50 | 1.369 | 0.794 | 1.077 | 3.000 | |
2000 | 30 | 1.281 | 0.789 | 1.599 | 2.357 | |
40 | 1.351 | 0.684 | 1.579 | 2.307 | |
50 | 1.399 | 0.806 | 1.754 | 2.346 | |
True Values | 1.350 | 0.750 | 1.680 | 2.320 | 0 |
Table 17.
Comparison of AO with AOA, SCA, and RSA against the true values for the CAR model at 0.08 noise variance.
Table 17.
Comparison of AO with AOA, SCA, and RSA against the true values for the CAR model at 0.08 noise variance.
Algorithm | Generations (T) | Population (Np) | Design Parameters | Best Fit |
---|
r1 | r2 | s1 | s2 |
---|
AO | 1000 | 30 | 1.357 | 0.772 | 1.754 | 2.271 | |
40 | 1.370 | 0.772 | 1.715 | 2.322 | |
50 | 1.364 | 0.771 | 1.712 | 2.318 | |
1500 | 30 | 1.380 | 0.784 | 1.737 | 2.337 | |
40 | 1.371 | 0.778 | 1.707 | 2.344 | |
50 | 1.372 | 0.776 | 1.722 | 2.328 | |
2000 | 30 | 1.367 | 0.776 | 1.741 | 2.312 | |
40 | 1.372 | 0.775 | 1.746 | 2.304 | |
50 | 1.358 | 0.776 | 1.763 | 2.272 | |
RSA | 1000 | 30 | 1.143 | 0.529 | 1.575 | 1.928 | |
40 | 1.148 | 0.604 | 2.135 | 1.461 | |
50 | 1.141 | 0.562 | 1.283 | 2.256 | |
1500 | 30 | 1.120 | 0.598 | 1.752 | 1.709 | |
40 | 1.277 | 0.690 | 1.318 | 2.498 | |
50 | 1.293 | 0.694 | 1.210 | 2.656 | |
2000 | 30 | 1.205 | 0.661 | 1.549 | 2.053 | |
40 | 1.323 | 0.706 | 1.275 | 2.556 | |
50 | 1.304 | 0.733 | 1.536 | 2.333 | |
SCA | 1000 | 30 | 1.416 | 0.831 | 1.623 | 2.574 | |
40 | 1.359 | 0.806 | 1.688 | 2.380 | |
50 | 1.350 | 0.738 | 1.705 | 2.289 | |
1500 | 30 | 1.364 | 0.801 | 1.602 | 2.486 | |
40 | 1.375 | 0.788 | 1.679 | 2.379 | |
50 | 1.317 | 0.755 | 1.695 | 2.255 | |
2000 | 30 | 1.344 | 0.773 | 1.623 | 2.366 | |
40 | 1.399 | 0.778 | 1.658 | 2.429 | |
50 | 1.340 | 0.713 | 1.667 | 2.250 | |
AOA | 1000 | 30 | 1.427 | 0.870 | 1.470 | 2.808 | |
40 | 1.430 | 0.811 | 1.295 | 2.854 | |
50 | 1.315 | 0.737 | 1.738 | 2.167 | |
1500 | 30 | 1.419 | 0.789 | 1.446 | 2.662 | |
40 | 1.339 | 0.779 | 1.702 | 2.328 | |
50 | 1.369 | 0.794 | 1.077 | 3.000 | |
2000 | 30 | 1.281 | 0.789 | 1.599 | 2.357 | |
40 | 1.351 | 0.684 | 1.579 | 2.307 | |
50 | 1.399 | 0.806 | 1.754 | 2.346 | |
True Values | 1.350 | 0.750 | 1.680 | 2.320 | 0 |
Table 18.
Comparison of AO with RSA, SCA, and AOA against average fit for the CAR model at 0.04 noise variance.
Table 18.
Comparison of AO with RSA, SCA, and AOA against average fit for the CAR model at 0.04 noise variance.
Generations (T) | Population (Np) | AO | RSA | AOA | SCA |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 19.
Comparison of AO with RSA, SCA, and AOA against average fit for the CAR model at 0.06 noise variance.
Table 19.
Comparison of AO with RSA, SCA, and AOA against average fit for the CAR model at 0.06 noise variance.
Generations (T) | Population (Np) | AO | RSA | AOA | SCA |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |
Table 20.
Comparison of AO with RSA, SCA, and AOA against average fit for the CAR model at 0.08 noise variance.
Table 20.
Comparison of AO with RSA, SCA, and AOA against average fit for the CAR model at 0.08 noise variance.
Generations (T) | Population (Np) | AO | RSA | AOA | SCA |
---|
1000 | 30 | | | | |
40 | | | | |
50 | | | | |
1500 | 30 | | | | |
40 | | | | |
50 | | | | |
2000 | 30 | | | | |
40 | | | | |
50 | | | | |