Forecasting of landslide disasters based on bionics algorithm (Part 1: Critical slip surface searching)

摘要

Two key issues for landslide forecasting include searching critical slip surfaces and determining their parameters. To solve the former aspect of this problem, a new critical slip surface search method based on an immunised continuous ant colony algorithm is proposed. Artificial immune system and evolutionary algorithms are combined with a continuous ant colony algorithm to form a new bionics algorithm (the immunised continuous ant colony algorithm) for continuous optimisation. The new bionics algorithm was then combined with a limit equilibrium analysis to form a new global optimisation algorithm to search critical slip surfaces. Finally, the proposed method is verified through five typical numerical examples and is compared with a continuous ant colony algorithm and previous studies in the literature. The results indicate that the applicability, validity and effectiveness of the new method are all improved over previous studies and over the continuous ant colony algorithm.