Accurate prediction method and system for horizontal displacement of concrete faced rockfill dam

[0031] In one or more embodiments, a method for accurately predicting the horizontal displacement of a concrete face rockfill dam based on statistical optimization neural network technology is disclosed.

[0032] To illustrate this example more clearly, refer to figure 1 , the implementation process of this method can be specifically described as follows:

[0033] (1) Acquisition of measured data. Select a concrete face rockfill dam as the application object, through long-term on-site monitoring of the target dam (if the dam has accumulated a large amount of measured data, it can be called directly), continuously monitor and accumulate the horizontal displacement of the dam body , dam body temperature, reservoir water level and other elements, it is required that the time series of the measured data of the above monitoring elements last for more than one year. If the dam has accumulated a large amount of measured data (time series over one year) including dam body horizontal di

[0074] In one or more embodiments, a precise system for predicting the horizontal displacement of a concrete face rockfill dam is disclosed, including:

[0075] A module for obtaining historical measured data of the concrete face rockfill dam to be tested and preprocessing the obtained data;

[0076] It is used to classify the environmental factors that affect the horizontal displacement of concrete face rockfill dams, and select the input variable module of the multiple linear regression model according to the correlation between each influencing factor and the horizontal displacement;

[0077] A module for establishing a multiple linear regression model for predicting the horizontal displacement of concrete face rockfill dams considering the delayed response of rockfill materials and the deformation characteristics of cyclic loading and unloading, and obtaining the preliminary predicted value of the horizontal displacement;

[0078] It is used to establish a statistically optim