Risk prediction method of pipe burst in water supply network based on Bayesian survival analysis

Abstract

The invention relates to a pipeline burst risk prediction method based on Bayesian survival analysis. The method comprises the following steps: (1) establishing a pipe burst database according to the collected pipe burst historical data, and extracting key information as a covariate; (2) Perform spatial clustering analysis on the pipe explosion points, and quantify the spatial distribution information of the pipe explosion points as a new covariate to supplement the pipe explosion database; (3) Based on the pipe explosion database, Bayesian survival is adopted. The analysis method is used to construct a pipe burst risk prediction model; (4) the pipe burst risk prediction model is used to predict the pipe burst risk of the pipeline. Compared with the prior art, the prediction result of the present invention is more accurate and reasonable.

Description

technical field [0001] The invention relates to a pipe burst risk prediction method for a water supply pipe network, in particular to a pipe burst risk prediction method for a water supply pipe network based on Bayesian survival analysis. Background technique [0002] As one of the important public infrastructures of the city, the water supply pipe network is the main artery of the whole city and undertakes the important task of delivering water to users. In the past ten years, the speed of urbanization in my country has continued to accelerate, the demand for urban water has increased sharply, and the scale of urban water supply pipe networks has increased day by day. However, there are still a series of problems in my country's water supply network, such as the laying time of some pipelines is too long, the phenomenon of aging is serious, and poor-quality pipes (such as gray cast iron pipes) occupy a large proportion of the pipe network. At the same time, in terms of safe...

AI Technical Summary

Problems solved by technology

But its disadvantages are: (1) The physical mechanism of pipe bursting is relatively complex and there are many factors causing pipe bursting, and the physical prediction model often cannot include all factors, and its application is difficult; (2) The data required for building the model Difficult or expensive to obtain

But its disadvantages are: (1) It needs a large amount of historical data of squibs for training; (2) The prediction results obtained by the purely data-driven model may be inconsistent with the actual observed results

Images