The ink preset algorithm based on the model optimized by chaotic bee colony

Combining with the least squares support vector machine (LSSVM) which has the features of small samples, nonlinear and high-dimensional, this paper proposes a model of ink preset for offset press based on LSSVM. According to the blindness of parameter choice in LSSVM, and the features of good optimization ability but easy to fall into local optimum and slow convergence at later evolution stage of artificial bee colony algorithm, a chaotic bee colony (CBC) algorithm was brought forwarded, and further was used to optimize the parameters in LSSVM. An ink preset model for offset press based on least squares support vector machine optimized by chaotic bee colony (CBC-LSSVM) was proposed in the paper to solve the adverse effects for the lateral ink flow due to the movement of the vibrator rollers. The experiment shows that presetting ink key based on CBC-LSSVM have higher precision than the ink preset system equipped by the Roland Uniset 70 newspaper Web press.