A Judgment Method of Gas Pipeline Network Leakage Level Based on Improved Evidence Fusion Algorithm

Abstract

The invention discloses a gas pipeline network leakage level judgment method based on an improved evidence fusion algorithm. The method uses gas pipeline network information measured by sensors to obtain multiple evidence sources for judging gas pipeline network leakage; for weak decision-making evidence, the The weight of the focal element with the highest mass value is set to 1, and the other focal elements are multiplied by their respective focal element weights, thereby effectively expanding the difference between the maximum trust focal element and other focal elements, making the evidence fusion result more reliable and the convergence speed fast. For the conflicting evidence, the average evidence is used to replace the conflicting evidence, and the evidence distance weight is used to limit the conflicting evidence, so as to reduce its influence on the result of evidence synthesis, so that the weak decision-making evidence can play its decision-making role. Using the invention can improve the accuracy of gas pipeline network leakage level decision-making.

Description

Technical field[0001]The present invention relates to the field of decision evidence fusion, and more particularly to a method of determining a gas pipeline network leakage based on improved evidence fusion algorithm.Background technique[0002]The method of fusion can be used for the gas pipeline leak level judgment. This scheme uses the leak information extracted by the pressure and flow of the gas line, and constructs evidence sources; these evidence sources have redundant, complementarity, and conflict, and require effective fusion treatment, thereby improving gas pipe networks. Leak diagnosis accuracy.[0003]The fusion of the evidence source can be implemented using Dempster-Shafer (DS) evidence theory. The theoretical focus of evidence is to use Dempster synthesis rules to integrate multi-source information and provide decision-making support. Conflict evidence in practical applications is generally existing. When the information is highly conflict, the combined results are often...

AI Technical Summary

Problems solved by technology

[0004] (1) When the evidence is completely conflicting, the denominator of the composition rule in the algorithm is O, and the DS composition formula cannot be used

[0005] (2) When the evidence is highly conflicting, the result may be inconsistent with the intuition

Yager believes that all conflicting evidence is unavailable, so the normalization factor in the DS combination rule is removed, and all conflicting probabilities are assigned to uncertain propositions. However, since all evidence conflicts are negated, the uncertainty of the composite result is increased. ; On the contrary, Smets believes that all conflicts are available, and the conflict evidence is preserved through the set theory, but the high reliability is gathered in the union focal element, which is not good for decision-making; Sun Quan et al. think that the global conflict is partially available, using evidence Credibility redistributes conflicts; Dubois et al. no longer study the problem of global conflicts, but consider how to solve the problem of local conflicts, assign local conflicts to the intersection and union of conflict focal elements, and fuse them in a certain proportion to reduce unreliable evidence Influence on Fusion Results

Although the algorithm improvement method from the perspective of local conflicts is simple and effective, the new combination rules often destroy the good characteristics of DS combination rules, and the combination results are usually affected by the order of evidence combination

Images