Concrete dam deformation combined forecasting model construction method based on ARIMA and PSO-ELM

Abstract

The invention provides a concrete dam deformation combined forecasting model construction method based on ARIMA and PSO-ELM, and aims at the characteristics of high nonlinearity and uncertainty causedby mutual influence of multiple factors in the dam deformation process, irregular chaotic characteristics caused by complex noise pollution and the like. An ensemble empirical mode (EEMD) is used tocarry out adaptive analysis and processing on a residual sequence of a displacement hybrid model, a particle swarm optimization (PSO) algorithm is used to optimize ELM and select an optimal input weight matrix and hidden layer deviation, and a PSOELM model is constructed to optimize a nonlinear high-frequency induction signal of the PSOELM model; meanwhile, fitting prediction is carried out on a low-frequency trend signal by means of an autoregressive integral moving average model (ARIMA), and a multi-scale deformation optimization combination forecasting model is established. Compared with atraditional model, the built model is higher in prediction precision, noise interference in the monitoring sequence can be overcome, the multi-scale characteristic of the dam monitoring sequence can be reflected, and the dam monitoring data time sequence can be analyzed and judged more clearly and comprehensively.

Description

technical field [0001] The invention relates to the technical field of dam operation safety monitoring and management, in particular to a method for constructing a combined prediction model of concrete dam deformation based on ARIMA and PSO-ELM. Background technique [0002] The dam is affected by many complex factors such as the external load environment during its service, and its local and overall safety performance gradually fades over time. The dam deformation is an important indicator for evaluating the active behavior of the dam, which reflects the In the dynamic evolution process under the dual coupling effect of external environmental load and internal dam material performance evolution, through the collection and arrangement of deformation monitoring data, in-depth excavation of deformation evolution law and chaotic signal processing of monitoring signals, a real-time prediction model is established, which is very useful for evaluating large It is of great signific...

AI Technical Summary

Problems solved by technology

However, in the process of signal decomposition, the EMD method adopts the cubic spline interpolation function to fit the extreme points of the signal to obtain the upper and lower envelopes, which has boundary effects and other problems, which limits the accuracy of the decomposed components. influence, which is not conducive to the analysis of signal change characteristics implicit in the signal

Images