A Neural Network: Family Competition Genetic Algorithm and Its Applications in Electromagnetic Optimization

<table>Best fitness plotted as a function of number of generations for (a) the TM polarization, (b) TE polarization, and (c) <svg height="11.65" id="M100" style="vertical-align:-0.04095pt" version="1.1" viewbox="0 0 63.3125 11.65" width="63.3125" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<g transform="matrix(.017,-0,0,-.017,.062,11.537)"><path d="M592 498l-30 -3q-13 63 -31 89q-13 19 -33.5 25.5t-75.5 6.5h-70v-493q0 -60 16 -75.5t86 -19.5v-28h-286v28q68 4 83.5 19.5t15.5 75.5v493h-62q-60 0 -83 -7t-33 -24q-12 -17 -32 -90h-29q10 116 12 180h20q10 -16 21 -20.5t34 -4.5h395q19 0 28.5 5t21.5 20h21
q2 -89 11 -177z" id="x54"></path></g><g transform="matrix(.017,-0,0,-.017,10.432,11.537)"><path d="M861 0h-263v28q61 5 73 22t12 97l-4 408h-4l-244 -548h-20l-224 534h-2l-15 -285q-5 -98 -3 -166q1 -34 19.5 -45t72.5 -17v-28h-233v28q45 5 62.5 17t22.5 43q11 65 19 183l14 208q3 36 3.5 55t-2.5 37.5t-6.5 25t-16.5 13t-25.5 8t-39.5 4.5v28h164l228 -501l232 501
h170v-28q-68 -6 -81 -21.5t-11 -91.5l9 -362q2 -80 15 -97t78 -22v-28z" id="x4D"></path></g><g transform="matrix(.017,-0,0,-.017,29.351,11.537)"><path d="M535 230h-212v-233h-58v233h-213v50h213v210h58v-210h212v-50z" id="x2B"></path></g><g transform="matrix(.017,-0,0,-.017,43.103,11.537)"><path d="M592 498l-30 -3q-13 63 -31 89q-13 19 -33.5 25.5t-75.5 6.5h-70v-493q0 -60 16 -75.5t86 -19.5v-28h-286v28q68 4 83.5 19.5t15.5 75.5v493h-62q-60 0 -83 -7t-33 -24q-12 -17 -32 -90h-29q10 116 12 180h20q10 -16 21 -20.5t34 -4.5h395q19 0 28.5 5t21.5 20h21
q2 -89 11 -177z" id="x54"></path></g><g transform="matrix(.017,-0,0,-.017,53.592,11.537)"><path d="M517 162l29 -4q-18 -100 -35 -158h-482v28q68 5 83.5 20t15.5 77v398q0 66 -13 80t-73 19v28h446q0 -8 8 -144l-29 -4q-11 51 -32 85q-11 17 -35 23t-78 6h-76q-23 0 -28 -5t-5 -27v-228h94q67 0 83 12t25 63h29v-191h-29q-10 54 -26 65.5t-81 11.5h-95v-189
q0 -65 24 -81q19 -13 104 -13q57 0 83 7t43 27q28 33 50 94z" id="x45"></path></g>
</svg> polarization absorber designs, respectively. In these cases, GA, FCGA, and NN-FCGA are all performed for comparison.</table>

Applied Computational Intelligence and Soft Computing

fig8

Figure 8

Figure 8: A Neural Network: Family Competition Genetic Algorithm and Its Applications in Electromagnetic Optimization 