A Danger-Theory-Based Immune Network Optimization Algorithm

Algorithm 2

The description of dt-aiNet.

1. Initialize. Randomly generate the initial network population within the definition domain, and set initial concentrations;

2. While (termination conditions are not meet) do

Begin

 2.1. Compute the affinity and danger signals of each antibody in the population;

 2.2. Select better individuals to clone, and make them active. The number of clones is related to concentrations;

 2.3. Perform the mutation operation to the clones, and then affinity mutation occurs. The mutation rate is related

 to affinities and can be adaptively adjusted;

 2.4. Perform the clonal suppression, and select better individuals to add into the network;

 2.5. Update the fitness, danger signals and concentrations of the population, and perform the network suppression;

 2.6. Randomly generate a certain number of antibodies, and add them into the network;

End;

3. Update the fitness, danger signals and concentrations of the population, and perform the network suppression;

4. Output the population.