Research Article
Radar Working State Recognition Based on Improved HPSO-BP
Algorithm 1
Pseudocode of improved HPSO-BP algorithm.
| (1) | Begin | | (2) | For () // is the particle population size | | (3) | Initialise velocity and position for particle ; | | (4) | Calculate the fitness value of particle and set ; // is the individual extreme value | | (5) End for | | (6) | ;// is the group extreme value | | (7) | For () // is the maximum number of iterations | | (8) | Improve , , and according to equations (6) and (7); | | (9) | For () | | (10) | Update and according to equations (1) and (2); | | (11) | Apply the crossover and mutation operator to particle according to equations (3)–(5); | | (12) | Calculate the fitness value of particle ; | | (13) | If () | | (14) | ; | | (15) | End if | | (16) | If () | | (17) | ; | | (18) | End if | | (19) | End for | | (20) | End for | | (21) | Determine the optimal weight and threshold of the BP neural network according to ; | | (22) | While (the training target error is not reached) | | (23) | Train the BP neural network; | | (24) | End while | | (25) | Predict the results on the test set; | | (26) | End |
|