Crossover probability

A standard GA is used in the experiments and the whole set of experiments was done on-computer (i.e., the actual programmable chip was not used during evolution). The GA parameters follow. Population size is three hundred selection method is fitness proportional with elitism crossover probability equals 0.8 (then 0 in successful runs) mutation probability is 0.1 the maximum number of generations was set to one hundred fifty. 

From the population, pairs of models are selected (randomly or with a probability proportional to their quality). Then, for each pair of models the common characteristics are preserved (i.e. variables excluded in both models remain excluded, variables included in both models remain included). For variables included in one model and excluded from the other, a random number is tried and compared with the crossover probability pc if the random number is lower than the cross-over probability, the excluded variable is included in the model and vice versa. Finally, the statistical parameter for the new model is calculated if the parameter value is better than the worst value in the population, the model is included in the population, in the place corresponding to its rank otherwise, it is no longer considered. This procedure is repeated for several pairs (for example 100 times). 

The Pareto-optimal solutions for the above optimization problem for nitrogen cooling are shown in Fig. 8.12. These results were obtained with NSGA-II after 400 generations with a population size of 100. The other GA parameters were crossover probability = 0.65, mutation probability = 0.25 and random number seed = 0.8. These values were chosen after around 15 trials with different values of the GA parameters, in order to get a very good spread of the Pareto-optimal solutions. 

Recombine P(i) with crossover probability Pc Mutate P(i) with mutation probability pm Evaluate P i) fitness end while... 

The elimination of water from the cathode GDL is facilitated by adding a coating of poly(tetrafluoroethylene) to the DL The hydrophobicity of PTFE can expell excess water from the GDL to the air/oxygen flowing inside the channels, avoiding flooding. Park et al. [46] have also shown that the MPL in the cathode DL could contribute to decrease methanol crossover, probably as a consequence of inducing hydraulic pressure on the cathode size [47]. 

In the crossover step, several individuals are selected in accordance with predefined crossover probability and are taken in pairs (parents) to exchange information and generate new individuals (children). Each pair of parents is able to create two children. Due to the fact that thresholds must be ordered within individuals, children s new values for k components depend on both n and parents current values. 

Number of chromosomes in the population 200 Number of generations (termination criterion) 1000 Selection Standard Roulette Replacement Children - Parents Mutation Probability 0.001 Crossover probability (one-site) 1... 

Step 1 Identify and enter the population size pop ize, crossover probability pc, mutation probability pm and termination generations max ge ... 

Step 1 First determine the number Pc pop size, of chromosome of the population which will crossover in the operation according to their crossover probability. 

In the hybrid intelligent algorithm, population size = 100, crossover probability Pc = 0.6, mutation probability = 0.5, number of iterations Gmax = 20,000, rank-based evaluation function a = 0.05 and there are 3000 random simulation. The main frequency of the PC for calculating is 2400 MHz, and all the algorithm program is realized by C++ language. 

In the hybrid intelligent algorithm, set population size N = 130, crossover probability p = 0.4, mutation probability = 0.5, number of iterations... 

Step 1 Identify and input population size pop size, crossover probability mutation probability and termination of generations max gcM. 

Step 1 First, determine the number of chromosomes Pc pop size that will be applied crossover operation for each population according to the crossover probability. 

In hybrid intelligent algorithm, confidence level j = 0.9, population size N = 90, crossover probability = 0.6, mutation probability = 0.5, iteration times Gmax = 30,000, in a ranking-based evaluation function a = 0.05, and the times of simulation loop are 6000. The frequency of the PC is 2400 MHz and the algorithm process is conducted in C++ programming language. 

The above simulation results indicate that for the algorithm with the same parameters, the more iteration times lead to the better optimization result. In addition to the number of iterations, three parameters including crossover probability, mutation probability and population size may also influence the accuracy of the solution to some degree. Therefore, appropriate parameter values should be set according to specific data condition when the algorithm is applied to solve the model. 

Determine and input the pop size, crossover probability Pc, mutation probability P, termination generations max gen and initial generations 1 = 0. 

In Hybrid Intelligent Algorithm, population size N = 60, crossover probability... 

In Hybrid Intelligent Algorithm, population size N = 70, crossover probability Pc = 0.5, mutation probability pm = 0.6, iteration times Gmax = 10,000, number of times for fuzzy stochastic simulation is 6000, with 3000 training samples. Figure 6.10 illustrates the convergence process of the objective function. 

---