Tunnel water leakage risk spatial dependency prediction method and prediction system

Abstract

The invention belongs to the technical field of shield tunnel water leakage risk dynamic sensing and control, and discloses a tunnel water leakage risk spatial dependence prediction method and prediction system, and the method comprises the steps: constructing a risk index system affecting shield tunnel water leakage through analyzing a water leakage risk causing mechanism; performing model design, including PCBN model Bayesian network structure design and risk index optimal edge distribution identification and fitting; performing model verification, wherein model application comprises water leakage risk evaluation and risk index correlation analysis. According to the invention, the dynamic reasoning and diagnosis of the Bayesian network model and the inter-index dependency processing advantages of the Copula theory are combined; according to the method, a Pair-Copula Bayesian model for shield tunnel water leakage risk assessment is established, and the occurrence and development mechanism of water leakage diseases in the actual shield tunnel operation period is fitted to the maximum extent, so a more accurate risk assessment result is obtained and serves as a risk management real-time decision support tool.

Description

technical field [0001] The invention belongs to the technical field of dynamic perception and control of shield tunnel water leakage risk, and in particular relates to a spatial dependence prediction method and prediction system for tunnel water leakage risk, in particular to a Pair-Copula Bayesian (PCBN) model Spatial dependency prediction method and system for water leakage risk in tunnels. Background technique [0002] At present, economic development has brought rapid development of rail transit, but due to the special structural characteristics and complex environment of shield tunnels, diseases frequently occur during the operation period of the tunnels. As one of the most common diseases, water leakage disease can not only directly lead to the damage of subway circuits and train stoppage, but also aggravate other types of tunnel diseases, and seriously threaten the safe operation of tunnels. Therefore, the risk assessment and management of shield tunnel water leakage...

AI Technical Summary

Problems solved by technology

[0003] The shield tunnel water leakage risk assessment method studied in the prior art mainly includes using the AHP method to determine the impact weight of the risk index on the system risk, constructing a fuzzy comprehensive evaluation matrix through the gray correlation coefficient method and using the multi-level weight distribution obtained by the rough set theory. Using the fuzzy comprehensive evaluation method to construct the leakage grade evaluation index system and hierarchical evaluation structure, etc., although the above methods have basically realized the qualitative and quantitative evaluation of water leakage risk, but there are many qualitative components and strong subjectivity , the calculation is complicated, and neither of them considers the dependence relationship between indicators

In order to solve the problem of information uncertainty of risk factors in complex systems, Bayesian Network (BN) has been applied to the risk assessment of tunnel water leakage, and the forward reasoning technology of BN is used to quantitatively predict the probability of water leakage disasters. Using BN reverse reasoning technology to carry out dynamic cycle diagnosis and determine the cause of leakage water damage in real time, but these analyzes mostly study the risk status of tunnel water leakage from the perspective of single, independent and static factors, and the results obtained are not accurate enough

[0004] Through the above

Images