A Novel Tournament Selection Based Differential Evolution Variant for Continuous Optimization Problems

Algorithm 1

Pseudocode of tournament selection based DE algorithm (TSDE).

    (1) Generate the initial population for generation and randomly initialize

    each population member where

    (2) FOR to NP

           Calculate fitness for each population member

       END FOR

    (3) WHILE the stopping criterion is not true

                    /^*Start of TSDE vectors selection ^*/

        Step  3.1.  TSDE vectors selection

           FOR to number of TSDE vectors

              FOR to Tournment_size

                  Select th tournament member with its fitness randomly from current population

              END FOR

              Select best of best member from the current tournament as   

              Return member index to be used as one of TSDE vectors in proposed mutation strategy

           END FOR

                    /^*End of TSDE vectors selection ^*/

     Step  3.2.  Mutation Step

              FOR to NP

               For the th target vector generate a donor vector with the

               specified mutation strategy (From Table 1 strategies or proposed TSDE strategy)

              END FOR

     Step  3.3.  Crossover Step

           FOR to NP

               For the th target vector generate a trial vector with the

               specified crossover scheme (Equation (2) or Equation (3))

           END FOR

    Step  3.4.  Selection Step

        FOR to NP

       Evaluate the trial vector against the target vector with fitness function f

       IF , THEN ,

                  IF , THEN ,

                  END IF

       END IF

        END FOR

    Step  3.5.  increment generation number

END WHILE