An intelligent prediction model of settlement deformation for the earth-rock dam

Abstract

The deformation prediction model is the key technical support of dam structure safety analysis. Given the shortcomings of the existing prediction models for settlement deformation of earth-rock dams in terms of accuracy and generalization, the adaptive network-based fuzzy inference system is introduced into the prediction model of earth-rock dam settlement deformation. The dynamic weight particle swarm optimization algorithm is used to optimize the parameters of the fitness value of the fuzzy layer in the adaptive network-based fuzzy inference system, the adaptive network-based fuzzy inference system that can find the optimal fitness value is formed, and the settlement deformation prediction model of earth-rock dam based on DPSO-ANFIS is established in this paper. The application of engineering examples can draw that the maximum deviation between the forecast output value of the settlement deformation prediction model based on DPSO-ANFIS and the measured value is 0.03mm, the root mean square error is 0.02mm, and the average absolute error is 0.01mm. The accuracy indexes are better than those of the settlement deformation prediction model of earth-rock dam based on PSO-ANFIS, ANFIS, and BP neural networks. For different measuring points and prediction interval length, the forecast output values of the settlement deformation prediction model of earth-rock dam based on DPSO-ANFIS is close to the measured values, the prediction trend conforms to the true value trend, the overall prediction performance is stable, and the accuracy value is within a reasonable range and better than other models. The application results of engineering examples show that the settlement deformation prediction model of earth-rock dam based on DPSO-ANFIS has high accuracy, good generalization, and reliable stability, and the comprehensive performance of engineering practice is better.