并行遗传算法源程序(主从式)(国外英语资料).docVIP

并行遗传算法源程序(主从式)(国外英文资料) Parallel genetic algorithm source (master-slave), the latest revision of 2009-02-25 11:06//, which is a simple parallel genetic algorithm source code, by the Denis Cormier (North Carolina State University) The serial / / program modification #define WIN32_LEAN_AND_MEAN / / exclude rarely used information from the Windows header #include stdio.h #include tchar.h #include stdlib.h #include malloc.h #include math.h #include iostream #include mpi.h #include windows.h #pragma comment (LIB, Wininet.lib) #pragma comment (LIB, mpi.lib) #pragma comment (LIB, cxx.lib) #define POPSIZE 6 / * the size of the population, is the total number of genes. #define MAXGENS 10 / * * / evolution #define NVARS 3 / * problem variables, i.e. the length of gene. #define PXOVER 0.8 / * * / crossover probability #define PMUTATION / * * / 0.15 mutation probability #define TRUE 1 #define FALSE 0 Int generation; the algebra has evolved the record / * * / Int cur_best; / * * / the best individual FILE *galog; / * * / log file Struct genotype / * * / gene type { Double gene[NVARS]; / * * / chromosome string Double upper[NVARS]; / * based on the upper bound of each Double lower[NVARS]; / * lower bound of each gene Double fitness; / * * / Fitness Double rfitness; / * * / relative fitness Double cfitness; / * * / cumulative fitness }; Struct genotype population[POPSIZE+1]; / * * / population Struct genotype newpopulation[POPSIZE+1]; / * * / new species Void initialize (void); / * * / population initialization function Double randval (double, double); / * * / random function Void keep_the_best (void); / * * / find the best individual Void elitist (void); / * * / maintain optimal Void select (void); / * * / selection operator Void crossover (void); / * * / crossover operator Void Xover (int, int); / * * / crossover operation Void swap (double *, double *); / * * / exchange Void evaluate (void); / * * / fitness evaluation Void worker (genotype *, int *) from the processor operation,