Intelligent Modeling for Transfer Function Control of DBR Semiconductor Laser at Near-Working Point

In this paper, a novel intelligent modeling method for transfer function control of DBR semiconductor lasers at near-working point is proposed, which is based on the quantum particle swarm optimization (QPSO) algorithm. This modeling method effectively solves the difficulty of application in the traditional laser theory and equivalent circuit models in practical engineering. First, we analyze the input and output characteristics of DBR semiconductor laser at near-working point, where the laser system can be equivalent to a transfer function model including two main modules: the laser power control module and the laser wavelength control module, and determine the structure of this equivalent model based on the analysis results. Then, we use the QPSO intelligent algorithm to identify the model parameters and finally obtain the equivalent transfer function model that can be easily applied to practical engineering. The standard deviations of the steady-state errors of these two modules are 3.4×10
^-3
and 1.2×10
^-5
, respectively. Experiments verify the effectiveness and convenience of the proposed (intelligent modeling) method, which can be used for on-line modeling of the DBR semiconductor laser at near-working point.