AO-BBO: A Novel Optimization Algorithm and Its Application in Plant Drug Extraction

Area of Science:

• Computational intelligence
• Optimization algorithms
• Biogeography-based optimization

Background:

• The standard biogeography-based optimization (BBO) algorithm exhibits limitations in search power under certain conditions.
• Addressing these limitations is crucial for improving the performance of BBO in complex optimization tasks.

Purpose of the Study:

• To propose an enhanced optimization algorithm, adaptive opposition-based biogeography-based optimization (AO-BBO), to overcome the shortcomings of the traditional BBO.
• To improve the global search capability and convergence accuracy of the BBO algorithm.

Main Methods:

• Developed the adaptive opposition-based biogeography-based optimization (AO-BBO) algorithm, integrating an opposition-based learning strategy.
• Implemented adaptive opposite learning probabilities based on the habitat suitability index (HSI) to prevent premature convergence.
• Validated the AO-BBO algorithm using nine benchmark functions.

Main Results:

• The AO-BBO algorithm demonstrated significant improvements in population diversity and global optimal solution search ability.
• Enhanced global exploration capability, convergence rate, and convergence accuracy were observed.
• Successfully applied the AO-BBO algorithm to optimize the parameters of a soft-sensing model for plant medicine extraction.

Conclusions:

• The AO-BBO algorithm effectively enhances the performance of the standard BBO.
• The optimized soft-sensing model achieved superior prediction accuracy and generalization ability.