Research on High-Precision Landslide Displacement Prediction Based on Improved Intelligent Optimization Algorithm and Machine Learning Model

The research on high-precision landslide displacement prediction has important reference value for landslide disaster prevention and warning. In this paper, the traditional Harris Hawks optimization (HHO) optimization algorithm is improved, while a high-precision landslide displacement prediction model (IHHO-BP) was established based on improved Harris Hawks Optimization (IHHO) algorithm combined with BP neural network. Combining the measured Beidou /GNSS data of Heifangtai landslide in China, the results show that the IHHO based on Levy variation, local enhancement and randomized Halton sequence population initialization strategy can effectively solve the problem that the traditional HHO algorithm has low search accuracy and will fall into local optimization. The combined prediction model of IHHO-BP has better generalization ability, and can effectively improve the prediction accuracy of landslide displacement. Taking the cumulative displacement prediction of a typical landslide as an example, the root mean square error of the displacement prediction of a typical landslide HF08 point by the IHO-BP combined prediction model is 2.6mm, which is higher than that of a single BP neural network (BP). Traditional HHO, Sparrow Search Algorithm (SSA), Particle Swarm Optimization (PSO) and Genetic Algorithm (Genetic Algorithm). GA and BP neural network combined prediction model (HHO-BP, SSA-BP, PSO-BP, GA-BP) improved the prediction accuracy by 68.29%, 13.33%, 51.85%, 18.75% and 35%, respectively. The root-mean-error of displacement prediction of IHO-BP model at HF05 point is 2.7mm, which is 37.21 %, 15.63%, 27.03%, 25% and 25% higher than that of BP, HHO-BP, SSA-BP, PSO-BP and GA-BP, respectively. The results show that the IHHO-BP landslide displacement combination prediction model constructed in this paper has higher prediction accuracy and better popularization and application value.