(1) Initialize the positions of particles with randomly generated binary values, and set velocities of particles to 0, the |
generation number , population size . |
(2) initialized Binary Population Set |
(3) |
(4) while Stopping criteria is not reached do |
(5) For each particle in do |
(6) Compute the fuzziness fitness functions and using the binary valued positions. |
(7) end for |
(8) Compare the particle’s current finesses and with particle’s best and . |
(9) if then |
(10) |
(11) end if |
(12) if then |
(13) |
(14) end if |
(15) Compare the global best fitness value with the population’s overall best values for the first fitness function, . |
(16) if then |
(17) |
(18) end if |
(19) Compare the global best fitness value with the population’s overall best values for the second fitness function, . |
(20) if then |
(21) |
(22) end if |
(23) Update the velocities of particles using Eq. (5) |
(24) Update the positions of particles using Eq. (6) |
(25) Insert the updated particles into the |
(26) Create the binary change-detection mask using Eq. (7) |
(27) |
(28) Use the Ranking and crowding distance operator [27] |
(29) Select Pareto solutions or non-dominated solutions |
(30) Insert selected Pareto solutions into the |
(31) |
(32) end while |
(33) Output1: Set of Pareto optimal solutions |
(34) Output2: Set of Binary change detection masks |