Construction and application of landslide displacement prediction model driven by semantic information
A technology of semantic information and prediction model, applied in semantic analysis, neural learning method, biological neural network model, etc., can solve the problem that the landslide disaster cannot distinguish between the flat stage and the acceleration stage, affects the prediction effect of landslide disaster, and the prediction of daily displacement is not accurate enough. and other problems to achieve the effect of improving the effect of landslide warning.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0047] This embodiment provides a landslide displacement prediction model driven by semantic information (hereinafter referred to as the landslide displacement prediction model), such as figure 1 As shown, the construction method of the prediction model is as follows:
[0048] S1. Through feature construction and feature combination, the landslide monitoring data is converted into semantic information, and individual combined samples and continuous combined samples are obtained
[0049] In a specific implementation, the landslide monitoring data is converted into semantic information through feature construction and feature combination, which is divided into the following two steps:
[0050] S1-1. Transform landslide monitoring data into semantic information features represented by text in a discretized manner
[0051] In a specific embodiment, the landslide monitoring data is exemplified by rainfall. Because landslide rainfall monitoring and the Meteorological Bureau use th...
Embodiment 2
[0115] This embodiment provides a landslide displacement prediction model driven by semantic information, which is constructed using the construction method in Embodiment 1, including:
[0116] The monitoring data semantic information conversion layer is used to convert landslide monitoring data into semantic information to obtain individual combined samples and continuous combined samples;
[0117] The phase division layer is used to divide the landslide cumulative displacement curve into gentle phase and acceleration phase;
[0118] The deep semantic feature extraction layer is used to extract the deep semantic features of continuous combination samples and individual combination samples according to the flat stage and the acceleration stage;
[0119] The stage identification and daily displacement prediction layer, including semantic information-driven gentle stage recognizer, used to identify the gentle stage and acceleration stage of the landslide; also includes semantic ...
Embodiment 3
[0121] This embodiment provides a landslide displacement prediction method driven by semantic information. According to landslide monitoring data, such as rainfall, using the landslide displacement prediction model driven by semantic information provided in Embodiment 2, the gentle stage and acceleration stage of the landslide Identify and predict the daily displacement during the acceleration phase of the landslide.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


