An improved immune algorithm for optimizing the pulse width modulation control sequence of inverters

ABSTRACT

In this article, an improved immune algorithm (IIA), based on the fundamental principles of the biological immune system, is proposed for optimizing the pulse width modulation (PWM) control sequence of a single-phase full-bridge inverter. The IIA takes advantage of the receptor editing and adaptive mutation mechanisms of the immune system to develop two operations that enhance the population diversity and convergence of the proposed algorithm. To verify the effectiveness and examine the performance of the IIA, 17 cases are considered, including fixed and disturbed resistances. Simulation results show that the IIA is able to obtain an effective PWM control sequence. Furthermore, when compared with existing immune algorithms (IAs), genetic algorithms (GAs), a non-traditional GA, simplified simulated annealing, and a generalized Hopfield neural network method, the IIA can achieve small total harmonic distortion (THD) and large magnitude. Meanwhile, a non-parametric test indicates that the IIA is significantly better than most comparison algorithms. Supplemental data for this article can be accessed at http://dx.doi.org/10.1080/0305215X.2016.1250894.

KEYWORDS:

• single-phase full-bridge inverter
• PWM control sequence
• immune algorithm
• adaptive mutation
• receptor editing

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The authors acknowledge support from the Scientific Research Foundation for Returned Overseas Chinese Scholars, State Education Ministry [Grant ZX20140265]; the Science Foundation of China University of Petroleum (Beijing) [Grant ZX20130080]; the National Natural Science Foundation of China [Grants 61304146 and 61473145]; the Provincial Science and Technology Foundation of Guizhou of China [Grant 20152002]; the Provincial Excellent Creative Talents of Science and Technology Reward Program of Guizhou of China [Grant 2014255].